diff --git a/aio_summary_robot.txt b/aio_summary_robot.txt deleted file mode 100644 index 2fbc40f..0000000 --- a/aio_summary_robot.txt +++ /dev/null @@ -1,12 +0,0 @@ -id aio -class_count 435 -class_count_with_text_definitions 414 -class_count_without_text_definitions 21 -distinct_synonym_count 471 -edge_count_by_predicate_rdfs:subClassOf 410 -non_deprecated_class_count 435 -rdf_triple_count 2205 -subclass_of_axiom_count 410 -synonym_statement_count 497 -synonym_statement_count_by_predicate_hasExactSynonym 361 -synonym_statement_count_by_predicate_hasRelatedSynonym 110 diff --git a/branch-counts.rq b/branch-counts.rq deleted file mode 100644 index be2c290..0000000 --- a/branch-counts.rq +++ /dev/null @@ -1,12 +0,0 @@ -prefix rdfs: -prefix owl: - -select ?branch (count(distinct ?child_class) as ?class_count) - where { - ?branch rdfs:subClassOf owl:Thing . - ?child_class rdfs:subClassOf* ?branch . -} -group by ?branch -order by ?branch - - diff --git a/is_a-edge-counts.rq b/is_a-edge-counts.rq deleted file mode 100644 index 9795321..0000000 --- a/is_a-edge-counts.rq +++ /dev/null @@ -1,21 +0,0 @@ -prefix rdfs: -prefix owl: -prefix oio: - -select distinct ?branch ?child_class ?is_a_edge - where { - values ?syn_type { - oio:hasExactSynonym - oio:hasRelatedSynonym - } - ?branch rdfs:subClassOf owl:Thing . - ?child_class rdfs:subClassOf* ?branch . - ?child_class rdfs:subClassOf ?is_a_edge . - minus { - ?child_class rdfs:subClassOf owl:Thing . - } - filter(isiri(?is_a_edge)) -} -order by ?branch ?child_class ?is_a_edge - - diff --git a/src/ontology/Makefile b/src/ontology/Makefile index 4eb44d3..6256cdf 100644 --- a/src/ontology/Makefile +++ b/src/ontology/Makefile @@ -700,4 +700,6 @@ Tricks: endef export data -include aio.Makefile \ No newline at end of file +include aio.Makefile + +make-aio: clean clean-extras remove-old-input components-from-new-input all \ No newline at end of file diff --git a/src/ontology/aio-src.csv b/src/ontology/aio-src.csv index acc05ff..bac06a0 100644 --- a/src/ontology/aio-src.csv +++ b/src/ontology/aio-src.csv @@ -1,449 +1,445 @@ -ID,Textual category,Category,Synomyms,Related synonyms,Name,description jaccard diff,Legacy Description,Claude opus Aristotelian definition,Reference (DOI),Informal layers list,Layers comment,layer parts,layer parts,Parent -ID,,AI oio:inSubset,A oio:hasExactSynonym SPLIT=|,A oio:hasRelatedSynonym SPLIT=|,LABEL,,,A IAO:0000115,>A oio:hasDbXref SPLIT=|,,A rdfs:comment,SC BFO:0000051 some % SPLIT=|,,SC % SPLIT=| -AIO:PReLULayer,Layer,AIO:LayerSubset,,,PReLU Layer,1.00,Parametric Rectified Linear Unit.,An activation layer that applies parametric rectified linear unit function element-wise.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/PReLU,,,,,AIO:ActivationLayer -AIO:ELULayer,Layer,AIO:LayerSubset,,,ELU Layer,0.85,Exponential Linear Unit.,An activation layer that applies the Exponential Linear Unit (ELU) function element-wise.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ELU,,,,,AIO:ActivationLayer -AIO:LeakyReLULayer,Layer,AIO:LayerSubset,,,LeakyReLU Layer,1.00,Leaky version of a Rectified Linear Unit.,An activation layer that applies the leaky rectified linear unit function element-wise.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/LeakyReLU,,,,,AIO:ActivationLayer -AIO:ReLULayer,Layer,AIO:LayerSubset,,,ReLU Layer,0.96,"Rectified Linear Unit activation function. With default values, it returns element-wise max(x, 0).",An activation layer that applies the rectified linear unit function element-wise.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ReLU,,,,,AIO:ActivationLayer -AIO:SoftmaxLayer,Layer,AIO:LayerSubset,,,Softmax Layer,0.92,Softmax activation function.,An activation layer that applies the softmax function to the inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Softmax,,,,,AIO:ActivationLayer -AIO:ThresholdedReLULayer,Layer,AIO:LayerSubset,,,ThresholdedReLU Layer,1.00,Thresholded Rectified Linear Unit.,An activation layer that applies the thresholded rectified linear unit function element-wise.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ThresholdedReLU,,,,,AIO:ActivationLayer -AIO:RankingBias,Bias,AIO:BiasSubset,,,Ranking Bias,0.33,"The idea that top-ranked results are the most relevant and important, leading to more clicks than other results.",An anchoring bias characterized by the idea that top-ranked results are the most relevant and important leading to more clicks than other results.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:AnchoringBias -AIO:DeepNeuralNetwork,Network,AIO:NetworkSubset,DNN,,Deep Neural Network,0.81,"A deep neural network (DNN) is a type of artificial neural network (ANN) characterized by multiple hidden layers between the input and output layers. Each layer consists of interconnected neurons that process and transmit information. DNNs can model complex patterns and representations in data through their hierarchical structure, where each layer extracts increasingly abstract features from the input. DNNs are widely used in various applications, including image and speech recognition, natural language processing, and more, due to their ability to learn and generalize from large amounts of data.",An artificial neural network characterized by multiple hidden layers between the input and output layers.,,,,,,AIO:ArtificialNeuralNetwork -AIO:FeedbackNetwork,Network,AIO:NetworkSubset,FBN,,Feedback Network,0.35,A feedback network iteratively refines its representations based on feedback from previous iterations' outputs.,An artificial neural network that refines its representations iteratively based on feedback from previous outputs.,,"Input, Hidden, Output, Hidden","Layers: Input, Hidden, Output, Hidden",,AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:ArtificialNeuralNetwork -AIO:Perceptron,Network,AIO:NetworkSubset,Single Layer Perceptron|SLP|Feed-Forward Network|FFN,,Perceptron,0.71,"A perceptron is a supervised learning algorithm for binary classification, deciding if an input belongs to a class using a linear predictor function that combines weights with the feature vector.",An artificial neural network with a supervised learning algorithm for binary classification using a linear predictor function.,,"Input, Output","Layers: Input, Output",,,AIO:ArtificialNeuralNetwork -AIO:MultiHeadAttentionLayer,Layer,AIO:LayerSubset,,,MultiHeadAttention Layer,0.96,"MultiHeadAttention layer. This is an implementation of multi-headed attention as described in the paper ""Attention is all you Need"" (Vaswani et al., 2017). If query, key, value are the same, then this is self-attention. Each timestep in query attends to the corresponding sequence in key, and returns a fixed-width vector.This layer first projects query, key and value. These are (effectively) a list of tensors of length num_attention_heads, where the corresponding shapes are (batch_size, , key_dim), (batch_size, , key_dim), (batch_size, , value_dim).Then, the query and key tensors are dot-producted and scaled. These are softmaxed to obtain attention probabilities. The value tensors are then interpolated by these probabilities, then concatenated back to a single tensor. Finally, the result tensor with the last dimension as value_dim can take an linear projection and return. When using MultiHeadAttention inside a custom Layer, the custom Layer must implement build() and call MultiHeadAttention's _build_from_signature(). This enables weights to be restored correctly when the model is loaded.",An attention layer that allows the model to attend to information from different representation subspaces.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/MultiHeadAttention,,,,,AIO:AttentionLayer -AIO:AdditiveAttentionLayer,Layer,AIO:LayerSubset,,,AdditiveAttention Layer,0.79,"Additive attention layer, a.k.a. Bahdanau-style attention.",An attention layer that implements additive attention also known as Bahdanau-style attention.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/AdditiveAttention,,,,,AIO:AttentionLayer -AIO:DenoisingAutoEncoder,Network,AIO:NetworkSubset,DAE|Denoising Autoencoder,,Denoising Auto Encoder,0.54,"Denoising autoencoders (DAEs) are neural networks trained to reconstruct the original undistorted input from a partially corrupted input, aiming to clean or denoise the corrupted input.",An autoencoder network trained to reconstruct the original undistorted input from a partially corrupted input.,https://doi.org/10.1145/1390156.1390294|,"Noisy Input, Hidden, Matched Output-Input","Layers: Noisy Input, Hidden, Matched Output-Input",,,AIO:AutoEncoderNetwork -AIO:DeepConvolutionalInverseGraphicsNetwork,Network,AIO:NetworkSubset,DCIGN,,Deep Convolutional Inverse Graphics Network,0.59,A deep convolutional inverse graphics network (DC-IGN) learns interpretable image representations disentangled for transformations like out-of-plane rotations and lighting variations. It consists of convolution and de-convolution layers and is trained using the stochastic gradient variational Bayes (SGVB) algorithm.,An autoencoder network that learns interpretable disentangled image representations through convolution and de-convolution layers trained with the stochastic gradient variational Bayes algorithm.,,"Input, Kernel, Convolutional/Pool, Probabilistic Hidden, Convolutional/Pool, Kernel, Output","Layers: Input, Kernel, Convolutional/Pool, Probabilistic Hidden, Convolutional/Pool, Kernel, Output",,,AIO:AutoEncoderNetwork -AIO:SparseAutoEncoder,Network,AIO:NetworkSubset,SAE|Sparse AE|Sparse Autoencoder,,Sparse Auto Encoder,0.66,"Sparse autoencoders have more hidden units than inputs but constrain only a few hidden units to be active at once, forcing the model to capture unique statistical features of the training data.",An autoencoder network with more hidden units than inputs that constrains only a few hidden units to be active at once.,,"Input, Hidden, Matched Output-Input","Layers: Input, Hidden, Matched Output-Input",,AIO:InputLayer|AIO:HiddenLayer|AIO:MatchedInputOutputLayer,AIO:AutoEncoderNetwork -AIO:VariationalAutoEncoder,Network,AIO:NetworkSubset,VAE,,Variational Auto Encoder,0.84,"A variational autoencoder (VAE) is a type of artificial neural network used for unsupervised learning. It consists of an encoder, which maps input data to a latent space, and a decoder, which reconstructs the input data from the latent space. Unlike traditional autoencoders, VAEs impose a probabilistic structure on the latent space, enabling them to generate new data samples by sampling from the learned latent distribution. This probabilistic approach allows VAEs to learn smooth and meaningful latent representations, making them useful for tasks such as data generation, anomaly detection, and semi-supervised learning.",An autoencoder network that imposes a probabilistic structure on the latent space for unsupervised learning.,,"Input, Probabilistic Hidden, Matched Output-Input","Layers: Input, Probabilistic Hidden, Matched Output-Input",,,AIO:AutoEncoderNetwork -AIO:BatchNorm1DLayer,Layer,AIO:LayerSubset,BatchNorm1d|BatchNorm1d|BatchNorm1D,,BatchNorm1D Layer,0.78,Applies Batch Normalization over a 2D or 3D input as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift .,A batch normalization layer that applies Batch Normalization over a 2D or 3D input.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:BatchNormalizationLayer -AIO:BatchNorm2DLayer,Layer,AIO:LayerSubset,BatchNorm2d|BatchNorm2d|BatchNorm2D,,BatchNorm2D Layer,0.87,Applies Batch Normalization over a 4D input (a mini-batch of 2D inputs with additional channel dimension) as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift .,A batch normalization layer that applies Batch Normalization over a 4D input.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:BatchNormalizationLayer -AIO:BatchNorm3DLayer,Layer,AIO:LayerSubset,BatchNorm3d|BatchNorm3d|BatchNorm3D,,BatchNorm3D Layer,0.87,Applies Batch Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension) as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift .,A batch normalization layer that applies Batch Normalization over a 5D input.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:BatchNormalizationLayer -AIO:LazyBatchNorm1DLayer,Layer,AIO:LayerSubset,LazyBatchNorm1d|LazyBatchNorm1d|LazyBatchNorm1D,,LazyBatchNorm1D Layer,0.77,A torch.nn.BatchNorm1d module with lazy initialization of the num_features argument of the BatchNorm1d that is inferred from the input.size(1).,A batch normalization layer that lazily initializes the num_features argument from the input size for 1D data.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:BatchNormalizationLayer -AIO:LazyBatchNorm2DLayer,Layer,AIO:LayerSubset,LazyBatchNorm2d|LazyBatchNorm2d|LazyBatchNorm2D,,LazyBatchNorm2D Layer,0.77,A torch.nn.BatchNorm2d module with lazy initialization of the num_features argument of the BatchNorm2d that is inferred from the input.size(1).,A batch normalization layer that lazily initializes the num_features argument from the input size for 2D data.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:BatchNormalizationLayer -AIO:LazyBatchNorm3DLayer,Layer,AIO:LayerSubset,LazyBatchNorm3d|LazyBatchNorm3d|LazyBatchNorm3D,,LazyBatchNorm3D Layer,0.77,A torch.nn.BatchNorm3d module with lazy initialization of the num_features argument of the BatchNorm3d that is inferred from the input.size(1).,A batch normalization layer that lazily initializes the num_features argument from the input size for 3D data.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:BatchNormalizationLayer -AIO:SyncBatchNormLayer,Layer,AIO:LayerSubset,SyncBatchNorm|SyncBatchNorm,,SyncBatchNorm Layer,0.95,Applies Batch Normalization over a N-Dimensional input (a mini-batch of [N-2]D inputs with additional channel dimension) as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift .,A batch normalization layer that applies synchronous Batch Normalization across multiple devices.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:BatchNormalizationLayer -AIO:SystemicBias,Bias,AIO:BiasSubset,Institutional Bias|Societal Bias,,Systemic Bias,0.15,Biases resulting from procedures and practices of particular institutions that operate in ways which result in certain social groups being advantaged or favored and others being disadvantaged or devalued.,A bias resulting from procedures and practices of institutions that operate in ways which result in certain social groups being advantaged or favored and others being disadvantaged or devalued.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:Bias -AIO:SocietalBias,Bias,AIO:BiasSubset,Social Bias,,Societal Bias,0.33,"Bias characterized by being for or against groups or individuals based on social identities, demographic factors, or immutable physical characteristics, often manifesting as stereotypes.",A bias characterized by being for or against groups or individuals based on social identities demographic factors or immutable physical characteristics often manifesting as stereotypes.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:Bias -AIO:ComputationalBias,Bias,AIO:BiasSubset,Statistical Bias,,Computational Bias,0.29,,A bias caused by differences between results and facts in the process of data analysis (including the source of data the estimator chose) and analysis methods.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:Bias -AIO:SunkCostFallacyBias,Bias,AIO:BiasSubset,Sunk Cost Fallacy,,Sunk Cost Fallacy Bias,0.40,"The tendency to continue an endeavor due to previously invested resources, despite costs outweighing benefits.",A bias characterized by the tendency to continue an endeavor due to previously invested resources despite costs outweighing benefits.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:Bias -AIO:HistoricalBias,Bias,AIO:BiasSubset,,,Historical Bias,0.69,"Long-standing biases encoded in society over time, distinct from biases in historical description or the interpretation of history, such as viewing the larger world from a Western or European perspective.",A bias characterized by long-standing biases encoded in society over time distinct from biases in historical description or interpretation.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:Bias -AIO:InstitutionalBias,Bias,AIO:BiasSubset,,,Institutional Bias,0.43,"Bias exhibited at the level of entire institutions, where practices or norms result in the favoring or disadvantaging of certain social groups, such as institutional racism or sexism.",A bias exhibited at the level of entire institutions where practices or norms result in the favoring or disadvantaging of certain social groups.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:Bias -AIO:IndividualBias,Bias,AIO:BiasSubset,,,Individual Bias,0.32,"A persistent point of view or limited list of such points of view applied by an individual, such as ""parent,"" ""academic,"" or ""professional.""",A persistent point of view or limited list of such points of view applied by an individual.,https://develop.consumerium.org/wiki/Individual_bias|GTP-4o with Seppala et al. 2017,,,,,AIO:Bias -AIO:HumanBias,Bias,AIO:BiasSubset,,,Human Bias,0.39,"Systematic errors in human thought based on heuristic principles, leading to simplified judgmental operations.",A systematic error in human thought based on heuristic principles leading to simplified judgmental operations.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:Bias -AIO:UnsupervisedBiclustering,Machine Learning,AIO:MachineLearningSubset,Block Clustering|Co-clustering|Two-mode Clustering|Two-way Clustering|Joint Clustering,,Unsupervised Biclustering,0.28,Methods that simultaneously cluster the rows and columns of an unlabeled input matrix to identify submatrices with coherent patterns.,A biclustering task focused on methods that simultaneously cluster the rows and columns of an unlabeled input matrix to identify submatrices with coherent patterns.,https://en.wikipedia.org/wiki/Biclustering|,,,,,AIO:Biclustering -AIO:SupervisedBiclustering,Machine Learning,AIO:MachineLearningSubset,Supervised Block Clustering|Supervised Co-clustering|Supervised Two-mode Clustering|Supervised Two-way Clustering|Supervised Joint Clustering,,Supervised Biclustering,0.36,"Methods that simultaneously cluster the rows and columns of a labeled matrix, considering data labels to enhance cluster coherence.",A biclustering task focused on methods that simultaneously cluster the rows and columns of a labeled matrix considering data labels to enhance cluster coherence.,https://en.wikipedia.org/wiki/Biclustering|,,,,,AIO:Biclustering -AIO:RestrictedBoltzmannMachine,Network,AIO:NetworkSubset,RBM,,Restricted Boltzmann Machine,0.35,A restricted Boltzmann machine (RBM) is a generative stochastic neural network that learns the probability distribution of its input data.,A Boltzmann machine network that learns the probability distribution of its input data.,https://en.wikipedia.org/wiki/Restricted_Boltzmann_machine,"Backfed Input, Probabilistic Hidden","Layers: Backfed Input, Probabilistic Hidden",,AIO:BackfedInputLayer|AIO:ProbabilisticHiddenLayer,AIO:BoltzmannMachineNetwork -AIO:CategoryEncodingLayer,Layer,AIO:LayerSubset,,,CategoryEncoding Layer,0.78,"A preprocessing layer which encodes integer features. This layer provides options for condensing data into a categorical encoding when the total number of tokens are known in advance. It accepts integer values as inputs, and it outputs a dense or sparse representation of those inputs. For integer inputs where the total number of tokens is not known, use tf.keras.layers.IntegerLookup instead.",A categorical features preprocessing layer that encodes integer features providing options for condensing data into a categorical encoding.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/CategoryEncoding,,,,,AIO:CategoricalFeaturesPreprocessingLayer -AIO:IntegerLookupLayer,Layer,AIO:LayerSubset,,,IntegerLookup Layer,0.25,A preprocessing layer which maps integer features to contiguous ranges.,A categorical features preprocessing layer that maps integer features to contiguous ranges.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/IntegerLookup,,,,,AIO:CategoricalFeaturesPreprocessingLayer -AIO:StringLookupLayer,Layer,AIO:LayerSubset,,,StringLookup Layer,0.25,A preprocessing layer which maps string features to integer indices.,A categorical features preprocessing layer that maps string features to integer indices.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/StringLookup,,,,,AIO:CategoricalFeaturesPreprocessingLayer -AIO:HashingLayer,Layer,AIO:LayerSubset,,,Hashing Layer,0.89,"A preprocessing layer which hashes and bins categorical features. This layer transforms categorical inputs to hashed output. It element-wise converts a ints or strings to ints in a fixed range. The stable hash function uses tensorflow::ops::Fingerprint to produce the same output consistently across all platforms. This layer uses FarmHash64 by default, which provides a consistent hashed output across different platforms and is stable across invocations, regardless of device and context, by mixing the input bits thoroughly. If you want to obfuscate the hashed output, you can also pass a random salt argument in the constructor. In that case, the layer will use the SipHash64 hash function, with the salt value serving as additional input to the hash function.",A categorical features preprocessing layer which hashes and bins categorical features.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Hashing,,,,,AIO:CategoricalFeaturesPreprocessingLayer -AIO:MulticlassClassification,Machine Learning,AIO:MachineLearningSubset,Multinomial Classification,,Multiclass Classification,0.44,Methods that classify instances into one of three or more classes.,A machine learning task focused on methods that classify instances into one of three or more classes.,https://en.wikipedia.org/wiki/Multiclass_classification|,,,,,AIO:Classification -AIO:HierarchicalClassification,Machine Learning,AIO:MachineLearningSubset,,,Hierarchical Classification,0.50,Methods that group things according to a hierarchy.,A classification task focused on methods that group things according to a hierarchy.,https://en.wikipedia.org/wiki/Hierarchical_classification|,,,,,AIO:Classification -AIO:DecisionTree,Machine Learning,AIO:MachineLearningSubset,,,Decision Tree,0.41,"A decision support tool that uses a tree-like model of decisions and their possible consequences, including chance event outcomes, resource costs, and utilities.",A machine learning model that uses a tree-like model of decisions and their possible consequences including chance event outcomes resource costs and utilities.,https://en.wikipedia.org/wiki/Decision_tree|,,,,,AIO:Classification -AIO:BinaryClassification,Machine Learning,AIO:MachineLearningSubset,,,Binary Classification,0.39,Methods that classify elements into two groups based on a classification rule.,A machine learning task focused on methods that classify elements into two groups based on a classification rule.,https://en.wikipedia.org/wiki/Binary_classification|,,,,,AIO:Classification -AIO:KnearestNeighborClassificationAlgorithm,Machine Learning,AIO:MachineLearningSubset,KNN|K-NN,,K-nearest Neighbor Classification Algorithm,,,"A classification and clustering that classifies objects by a plurality vote of its neighbors, assigning each object to the class most common among its k nearest neighbors.",https://en.wikipedia.org/wiki/K-nearest_neighbors_algorithm|,,,,,AIO:Classification|AIO:Clustering -AIO:UnsupervisedClustering,Machine Learning,AIO:MachineLearningSubset,Cluster analysis,,Unsupervised Clustering,0.26,Methods that group a set of unlabeled objects such that objects in the same group are more similar to each other than to those in other groups.,A clustering task focused on methods that group a set of unlabeled objects such that objects in the same group are more similar to each other than to those in other groups.,https://en.wikipedia.org/wiki/Cluster_analysis|,,,,,AIO:Clustering -AIO:SupervisedClustering,Machine Learning,AIO:MachineLearningSubset,Cluster analysis,,Supervised Clustering,0.38,"Methods that group labeled objects such that objects in the same group have similar labels, relative to those in other groups.",A clustering task focused on methods that group labeled objects such that objects in the same group have similar labels relative to those in other groups.,https://en.wikipedia.org/wiki/Cluster_analysis|,,,,,AIO:Clustering -AIO:HierarchicalClustering,Machine Learning,AIO:MachineLearningSubset,HCL,,Hierarchical Clustering,0.55,Methods that build a hierarchy of clusters.,A clustering method that builds a hierarchy of clusters.,https://en.wikipedia.org/wiki/Hierarchical_clustering|,,,,,AIO:Clustering -AIO:DunningKrugerEffectBias,Bias,AIO:BiasSubset,Dunning-Kruger Effect,,Dunning-Kruger Effect Bias,,,A cognitive bias in which people with low ability in an area overestimate that ability. Often measured by comparing self-assessment with objective performance.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:CognitiveBias -AIO:UseAndInterpretationBias,Bias,AIO:BiasSubset,Interpretive Bias,,Use And Interpretation Bias,0.67,"Bias inappropriately analyzing ambiguous stimuli, scenarios, and events.",A computational bias characterized by inappropriately analyzing ambiguous stimuli scenarios and events.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:ComputationalBias -AIO:SelectionAndSamplingBias,Bias,AIO:BiasSubset,Selection Bias|Sampling Bias|Selection Effect,,Selection And Sampling Bias,0.50,"Bias introduced by non-random selection of individuals, groups, or data, failing to ensure representativeness.",A computational bias introduced by non-random selection of individuals groups or data failing to ensure representativeness.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:ComputationalBias -AIO:ProcessingBias,Bias,AIO:BiasSubset,Validation Bias,,Processing Bias,0.59,"Judgment modulated by affect, influenced by the level of efficacy and efficiency in information processing; often referred to as aesthetic judgment in cognitive sciences.",A computational bias resulting from judgment modulated by affect influenced by the level of efficacy and efficiency in information processing.,https://en.wikipedia.org/wiki/Bias_(statistics)|GTP-4o with Seppala et al. 2017,,,,,AIO:ComputationalBias -AIO:Convolution1DLayer,Layer,AIO:LayerSubset,Conv1D Layer|nn.Conv1d|Conv1d|Convolution1D|Convolution1d,,Convolution1D Layer,0.33,1D convolution layer (e.g. temporal convolution).,A layer that implements 1D convolution (e.g. temporal convolution).,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv1D,,,,,AIO:ConvolutionalLayer -AIO:Convolution1DTransposeLayer,Layer,AIO:LayerSubset,Conv1DTranspose Layer|nn.ConvTranspose1d|ConvTranspose1d|Convolution1dTranspose|Convolution1DTranspose,,Convolution1DTranspose Layer,0.94,"Transposed convolution layer (sometimes called Deconvolution). The need for transposed convolutions generally arises from the desire to use a transformation going in the opposite direction of a normal convolution, i.e., from something that has the shape of the output of some convolution to something that has the shape of its input while maintaining a connectivity pattern that is compatible with said convolution. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 3) for data with 128 time steps and 3 channels.",A layer that implements transposed 1D convolution sometimes called deconvolution.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv1DTranspose,,,,,AIO:ConvolutionalLayer -AIO:Convolution2DLayer,Layer,AIO:LayerSubset,Conv2D Layer|nn.Conv2d|Conv2d|Convolution2D|Convolution2d,,Convolution2D Layer,0.90,"2D convolution layer (e.g. spatial convolution over images). This layer creates a convolution kernel that is convolved with the layer input to produce a tensor of outputs. If use_bias is True, a bias vector is created and added to the outputs. Finally, if activation is not None, it is applied to the outputs as well. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 128, 3) for 128x128 RGB pictures in data_format=""channels_last"". You can use None when a dimension has variable size.",A layer that implements 2D convolution (e.g. spatial convolution over images).,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv2D,,,,,AIO:ConvolutionalLayer -AIO:Convolution2DTransposeLayer,Layer,AIO:LayerSubset,Conv2DTranspose Layer|nn.ConvTranspose2d|ConvTranspose2d|Convolution2DTranspose|Convolution2dTranspose,,Convolution2DTranspose Layer,0.82,Transposed convolution layer (sometimes called Deconvolution).,A layer that implements transposed 2D convolution,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv2DTranspose,,,,,AIO:ConvolutionalLayer -AIO:Convolution3DLayer,Layer,AIO:LayerSubset,Conv3D Layer|nn.Conv3d|Conv3d|Convolution3D|Convolution3d,,Convolution3D Layer,0.88,"3D convolution layer (e.g. spatial convolution over volumes). This layer creates a convolution kernel that is convolved with the layer input to produce a tensor of outputs. If use_bias is True, a bias vector is created and added to the outputs. Finally, if activation is not None, it is applied to the outputs as well. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 128, 128, 1) for 128x128x128 volumes with a single channel, in data_format=""channels_last"".",A layer that implements 3D convolution (e.g. spatial convolution over volumes).,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv3D,,,,,AIO:ConvolutionalLayer -AIO:Convolution3DTransposeLayer,Layer,AIO:LayerSubset,Conv3DTranspose Layer|nn.ConvTranspose3d|ConvTranspose3d|Convolution3DTranspose|Convolution3dTranspose,,Convolution3DTranspose Layer,0.95,"Transposed convolution layer (sometimes called Deconvolution). The need for transposed convolutions generally arises from the desire to use a transformation going in the opposite direction of a normal convolution, i.e., from something that has the shape of the output of some convolution to something that has the shape of its input while maintaining a connectivity pattern that is compatible with said convolution. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 128, 128, 3) for a 128x128x128 volume with 3 channels if data_format=""channels_last"".",A layer that implements transposed 3D convolution,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv3DTranspose,,,,,AIO:ConvolutionalLayer -AIO:SeparableConvolution1DLayer,Layer,AIO:LayerSubset,SeparableConv1D Layer,,SeparableConvolution1D Layer,0.84,"Depthwise separable 1D convolution. This layer performs a depthwise convolution that acts separately on channels, followed by a pointwise convolution that mixes channels. If use_bias is True and a bias initializer is provided, it adds a bias vector to the output. It then optionally applies an activation function to produce the final output.a",A layer that performs depthwise separable 1D convolution.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/SeparableConv1D,,,,,AIO:ConvolutionalLayer -AIO:SeparableConvolution2DLayer,Layer,AIO:LayerSubset,SeparableConv2D Layer,,SeparableConvolution2D Layer,0.94,"Depthwise separable 2D convolution. Separable convolutions consist of first performing a depthwise spatial convolution (which acts on each input channel separately) followed by a pointwise convolution which mixes the resulting output channels. The depth_multiplier argument controls how many output channels are generated per input channel in the depthwise step. Intuitively, separable convolutions can be understood as a way to factorize a convolution kernel into two smaller kernels, or as an extreme version of an Inception block.",A layer that performs depthwise separable 2D convolution.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/SeparableConv2D,,,,,AIO:ConvolutionalLayer -AIO:ConvLSTM1DLayer,Layer,AIO:LayerSubset,,,ConvLSTM1D Layer,0.65,"1D Convolutional LSTM. Similar to an LSTM layer, but the input transformations and recurrent transformations are both convolutional.",A convolutional layer that implements a 1D Convolutional LSTM similar to an LSTM but with convolutional input and recurrent transformations.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ConvLSTM1D,,,,,AIO:ConvolutionalLayer -AIO:ConvLSTM2DLayer,Layer,AIO:LayerSubset,,,ConvLSTM2D Layer,0.65,"2D Convolutional LSTM. Similar to an LSTM layer, but the input transformations and recurrent transformations are both convolutional.",A convolutional layer that implements a 2D Convolutional LSTM similar to an LSTM but with convolutional input and recurrent transformations.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ConvLSTM2D,,,,,AIO:ConvolutionalLayer -AIO:ConvLSTM3DLayer,Layer,AIO:LayerSubset,,,ConvLSTM3D Layer,0.65,"3D Convolutional LSTM. Similar to an LSTM layer, but the input transformations and recurrent transformations are both convolutional.",A convolutional layer that implements a 3D Convolutional LSTM similar to an LSTM but with convolutional input and recurrent transformations.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ConvLSTM3D,,,,,AIO:ConvolutionalLayer -AIO:DepthwiseConv1DLayer,Layer,AIO:LayerSubset,,,DepthwiseConv1D Layer,0.95,"Depthwise 1D convolution. Depthwise convolution is a type of convolution in which each input channel is convolved with a different kernel (called a depthwise kernel). You can understand depthwise convolution as the first step in a depthwise separable convolution. It is implemented via the following steps: Split the input into individual channels. Convolve each channel with an individual depthwise kernel with depth_multiplier output channels. Concatenate the convolved outputs along the channels axis. Unlike a regular 1D convolution, depthwise convolution does not mix information across different input channels. The depth_multiplier argument determines how many filter are applied to one input channel. As such, it controls the amount of output channels that are generated per input channel in the depthwise step.",A layer that performs depthwise 1D convolution,https://www.tensorflow.org/api_docs/python/tf/keras/layers/DepthwiseConv1D,,,,,AIO:ConvolutionalLayer -AIO:DepthwiseConv2DLayer,Layer,AIO:LayerSubset,,,DepthwiseConv2D Layer,0.89,Depthwise 2D convolution.,A layer that performs depthwise 2D convolution,https://www.tensorflow.org/api_docs/python/tf/keras/layers/DepthwiseConv2D,,,,,AIO:ConvolutionalLayer -AIO:DataAugmentation,Preprocessing,AIO:PreprocessingSubset,Data Expansion|Data Enrichment,Paraphrasing|Synonym replacement,Data Augmentation,0.24,"A technique used to increase the diversity and quantity of training data by applying various transformations such as rotation, scaling, flipping, and cropping to existing data samples, enhancing the robustness and performance of machine learning models.",A technique used to increase the diversity and quantity of training data by applying various transformations such as rotation scaling flipping and cropping to existing data samples enhancing the robustness and performance of machine learning models.,,,,,,AIO:DataEnhancement -AIO:FeatureExtraction,Preprocessing,AIO:PreprocessingSubset,Attribute Extraction|Feature Isolation,Syntactic information|Semantic embeddings,Feature Extraction,0.07,"The process of transforming raw data into a set of measurable characteristics that can be used as input for machine learning algorithms, enhancing the ability to make accurate predictions.",The process of transforming raw data into a set of measurable characteristics that can be used as input for machine learning algorithms enhancing the ability to make accurate predictions.,,,,,,AIO:DataEnhancement -AIO:SubwordSegmentation,Preprocessing,AIO:PreprocessingSubset,Fragmentation|Part-word Division,Byte Pair Encoding|SentencePiece,Subword Segmentation,0.12,"The process of dividing text into subword units, which are smaller than words but larger than individual characters, to improve the efficiency and effectiveness of natural language processing models by capturing meaningful subunits of words.",The process of dividing text into subword units which are smaller than words but larger than individual characters to improve the efficiency and effectiveness of natural language processing models by capturing meaningful subunits of words.,,,,,,AIO:DataPreparation -AIO:Cleaning,Preprocessing,AIO:PreprocessingSubset,Data Cleansing|Standardization,Data cleaning|Text normalization,Cleaning,0.17,"The process of removing noise, inconsistencies, and irrelevant information from data to enhance its quality and prepare it for analysis or further processing.",The process of removing noise inconsistencies and irrelevant information from data to enhance its quality and prepare it for analysis or further processing.,,,,,,AIO:DataPreparation -AIO:Tokenization,Preprocessing,AIO:PreprocessingSubset,Lexical Analysis|Text Segmentation,,Tokenization,0.21,"The process of converting a sequence of text into smaller, meaningful units called tokens, typically words or subwords, for the purpose of analysis or processing by language models.",The process of converting a sequence of text into smaller meaningful units called tokens typically words or subwords for the purpose of analysis or processing by language models.,,,,,,AIO:DataPreparation -AIO:VocabularyReduction,Preprocessing,AIO:PreprocessingSubset,Vocabulary Condensation|Lexical Simplification|Lexicon Pruning,,Vocabulary Reduction,0.11,"The technique of limiting the number of unique tokens in a language model’s vocabulary by merging or eliminating less frequent tokens, thereby optimizing computational efficiency and resource usage.",The technique of limiting the number of unique tokens in a language model's vocabulary by merging or eliminating less frequent tokens thereby optimizing computational efficiency and resource usage.,,,,,,AIO:DataPreparation -AIO:Normalization,Preprocessing,AIO:PreprocessingSubset,,,Normalization,,"The technique of transforming data into a standard format or scale, typically to reduce redundancy and improve consistency, often involving the adjustment of values measured on different scales to a common scale.",The technique of transforming data into a standard format or scale typically to reduce redundancy and improve consistency often involving the adjustment of values measured on different scales to a common scale.,,,,,,AIO:DataPreparation -AIO:RadialBasisNetwork,Network,AIO:NetworkSubset,RBN|Radial Basis Function Network|RBFN,,Radial Basis Network,0.55,"Radial basis function networks use radial basis functions as activation functions, effective for pattern recognition and interpolation.",A deep feedforward network that uses radial basis functions as activation functions for pattern recognition and interpolation.,https://en.wikipedia.org/wiki/Radial_basis_function_network,"Input, Hidden, Output","Layers: Input, Hidden, Output",,AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:DeepFeedForwardNetwork -AIO:NeuralTuringMachineNetwork,Network,AIO:NetworkSubset,NTM,,Neural Turing Machine Network,0.66,"A neural Turing machine (NTM) combines neural network pattern matching with the algorithmic power of programmable computers, using attention mechanisms to interact with external memory for tasks like copying, sorting, and associative recall.",A deep feedforward network that combines neural network pattern matching with the algorithmic power of programmable computers.,https://en.wikipedia.org/wiki/Neural_Turing_machine,"Input, Hidden, Spiking Hidden, Output","Layers: Input, Hidden, Spiking Hidden, Output",,AIO:InputLayer|AIO:HiddenLayer|AIO:SpikingHiddenLayer|AIO:OutputLayer,AIO:DeepFeedForwardNetwork|AIO:LongShortTermMemory -AIO:RecursiveLanguageModel,Model,AIO:ModelSubset,RLM,Compositional generalization,Recursive Language Model,0.14,"A recursive language model uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions, improving systematic generalization abilities.",A language model that uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions improving systematic generalization abilities.,https://en.wikipedia.org/wiki/Recurrent_neural_network,"Input, Memory Cell, Output","Layers: Input, Memory Cell, Output",,,AIO:DeepNeuralNetwork -AIO:DeepConvolutionalNetwork,Network,AIO:NetworkSubset,DCN|Convolutional Neural Network|CNN|ConvNet,,Deep Convolutional Network,0.58,"A deep convolutional network (CNN) is an artificial neural network used to analyze visual imagery, utilizing shared-weight architecture and translation-equivariant feature maps.",A deep neural network specialized for analyzing visual imagery using shared-weight architecture and translation-equivariant feature maps.,https://en.wikipedia.org/wiki/Convolutional_neural_network,"Input, Kernel, Convolutional/Pool, Hidden, Output","Layers: Input, Kernel, Convolutional/Pool, Hidden, Output",,AIO:InputLayer,AIO:DeepNeuralNetwork -AIO:DeepActiveLearning,Network,AIO:NetworkSubset,DeepAL,,Deep Active Learning,0.33,Combining deep learning and active learning to maximize model performance gain while annotating the fewest samples possible.,A deep neural network that combines deep learning and active learning to maximize model performance while annotating the fewest samples possible.,https://arxiv.org/pdf/2009.00236.pdf,,,,,AIO:DeepNeuralNetwork -AIO:DeepFeedForwardNetwork,Network,AIO:NetworkSubset,DFF|Multilayer Perceptoron|MLP,,Deep Feed-Forward Network,,,A deep neural network that processes information in one direction—from input nodes through hidden nodes to output nodes—without cycles or loops.,https://en.wikipedia.org/wiki/Feedforward_neural_network,"Input, Hidden, Output","Layers: Input, Hidden, Output",,AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:DeepNeuralNetwork -AIO:MetricLearning,Network,AIO:NetworkSubset,Distance Metric Learning,,Metric Learning,0.47,Learning a representation function that maps objects into an embedded space.,A deep neural network that learns a representation function mapping objects into an embedded space.,https://paperswithcode.com/task/metric-learning,,,,,AIO:DeepNeuralNetwork -AIO:DeconvolutionalNetwork,Network,AIO:NetworkSubset,DN,,Deconvolutional Network,0.78,Deconvolutional networks allow unsupervised construction of hierarchical image representations for tasks such as denoising and feature extraction for object recognition.,A deep neural network that uses deconvolution for unsupervised construction of hierarchical image representations.,https://ieeexplore.ieee.org/document/5539957,"Input, Kernel, Convolutional/Pool, Output","Layers: Input, Kernel, Convolutional/Pool, Output",,,AIO:DeepNeuralNetwork -AIO:RepresentationLearning,Network,AIO:NetworkSubset,Feature Learning,,Representation Learning,0.41,Discovering representations required for feature detection or classification from raw data.,A deep neural network that discovers representations required for feature detection or classification from raw data.,https://en.wikipedia.org/wiki/Feature_learning,,,,,AIO:DeepNeuralNetwork -AIO:GraphConvolutionalNetwork,Network,AIO:NetworkSubset,GCN,,Graph Convolutional Network,0.68,"A graph convolutional network (GCN) operates directly on graph structures, utilizing their structural information for tasks like node classification and graph clustering.",A deep neural network that operates directly on graph structures utilizing structural information.,https://arxiv.org/abs/1609.02907,"Input, Hidden, Hidden, Output","Layers: Input, Hidden, Hidden, Output",,AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:DeepNeuralNetwork -AIO:GeneralizedFewshotLearning,Network,AIO:NetworkSubset,GFSL,,Generalized Few-shot Learning,,,"A deep neural network that learns novel classes from few samples per class, preventing catastrophic forgetting of base classes and ensuring classifier calibration.",https://paperswithcode.com/paper/generalized-and-incremental-few-shot-learning/review/,,,,,AIO:DeepNeuralNetwork -AIO:IncremenetalFewshotLearning,Network,AIO:NetworkSubset,IFSL,,Incremenetal Few-shot Learning,,,"A deep neural network trained on a base set of classes and then presented with novel classes, each with few labeled examples.",https://arxiv.org/abs/1810.07218,,,,,AIO:DeepNeuralNetwork -AIO:ContinualLearning,Network,AIO:NetworkSubset,Incremental Learning|Life-Long Learning,,Continual Learning,0.48,"Learning a model for sequential tasks without forgetting knowledge from preceding tasks, with no access to old task data during new task training.",A deep neural network that learns sequential tasks without forgetting knowledge from preceding tasks and without access to old task data during new task training.,https://paperswithcode.com/task/continual-learning,,,,,AIO:DeepNeuralNetwork -AIO:OneshotLearning,Network,AIO:NetworkSubset,OSL,,One-shot Learning,,,A deep neural network that classified objects from one or only a few examples.,https://en.wikipedia.org/wiki/One-shot_learning,,,,,AIO:DeepNeuralNetwork -AIO:RecursiveNeuralNetwork,Network,AIO:NetworkSubset,RecuNN|RvNN,,Recursive Neural Network,0.33,A recursive neural network applies the same set of weights recursively over structured input to generate structured or scalar predictions.,A deep neural network that recursively applies weights over structured input to generate structured or scalar predictions.,https://en.wikipedia.org/wiki/Recursive_neural_network,,,,,AIO:DeepNeuralNetwork -AIO:ResidualNeuralNetwork,Network,AIO:NetworkSubset,ResNN|Deep Residual Network|ResNet|DRN,,Residual Neural Network,0.35,"A residual neural network (ResNet) employs skip connections to bypass certain layers, facilitating the learning of residual functions.",A deep neural network that employs skip connections to bypass layers facilitating learning of residual functions.,https://en.wikipedia.org/wiki/Residual_neural_network,"Input, Weight, BN, ReLU, Weight, BN, Addition, ReLU","Layers: Input, Weight, BN, ReLU, Weight, BN, Addition, ReLU",,,AIO:DeepNeuralNetwork -AIO:ZeroshotLearning,Network,AIO:NetworkSubset,ZSL,,Zero-shot Learning,,,A deep neural network that predicts classes at test time from classes not observed during training.,https://en.wikipedia.org/wiki/Zero-shot_learning,,,,,AIO:DeepNeuralNetwork -AIO:ContrastiveLearning,Network,AIO:NetworkSubset,,,Contrastive Learning,0.82,"Contrastive learning is a self-supervised learning approach in which the model learns to distinguish between similar and dissimilar pairs of data samples. By maximizing the similarity between positive pairs (similar samples) and minimizing the similarity between negative pairs (dissimilar samples), the model learns to capture meaningful representations of the data. This method is particularly effective for representation learning and is widely used in tasks such as image classification, clustering, and retrieval. Contrastive learning techniques often employ loss functions such as the contrastive loss or the triplet loss to achieve these objectives.",A deep neural network self-supervised learning approach that learns to distinguish between similar and dissimilar data samples.,https://arxiv.org/abs/2202.14037,,,,,AIO:DeepNeuralNetwork -AIO:MultimodalDeepLearning,Network,AIO:NetworkSubset,,,Multimodal Deep Learning,0.63,Creating models that process and link information using various modalities.,A deep neural network that processes and links information using various modalities.,https://arxiv.org/abs/2105.11087,,,,,AIO:DeepNeuralNetwork -AIO:DeepTransferLearning,Network,AIO:NetworkSubset,,,Deep Transfer Learning,0.29,Relaxing the hypothesis that training data must be independent and identically distributed (i.i.d.) with test data to address insufficient training data.,A deep neural network that relaxes the hypothesis that training data must be independent and identically distributed with test data to address insufficient training data.,https://arxiv.org/abs/1808.01974,,,,,AIO:DeepNeuralNetwork -AIO:TransformerNetwork,Network,AIO:NetworkSubset,,,Transformer Network,0.63,"A transformer network utilizes attention mechanisms to weigh the significance of each part of the input data, widely used in natural language processing (NLP) and computer vision (CV).",A deep neural network that utilizes attention mechanisms to weigh the significance of input data.,https://en.wikipedia.org/wiki/Transformer_(machine_Learning_model),,,,,AIO:DeepNeuralNetwork -AIO:FederatedLearning,Network,AIO:NetworkSubset,,,Federated Learning,0.41,Training an algorithm across multiple decentralized edge devices or servers holding local data samples without exchanging them.,A deep neural network trained across decentralized edge devices or servers holding local data samples without exchanging them.,https://en.wikipedia.org/wiki/Federated_learning,,,,,AIO:DeepNeuralNetwork -AIO:MultidimensionalScaling,Machine Learning,AIO:MachineLearningSubset,MDS,,Multidimensional Scaling,0.08,A method that translates information about the pairwise distances among a set of objects or individuals into a configuration of points mapped into an abstract Cartesian space.,A dimensionality reduction method that translates information about the pairwise distances among a set of objects or individuals into a configuration of points mapped into an abstract Cartesian space.,https://en.wikipedia.org/wiki/Multidimensional_scaling|,,,,,AIO:DimensionalityReduction -AIO:PrincipalComponentAnalysis,Machine Learning,AIO:MachineLearningSubset,PCA,,Principal Component Analysis,0.32,"A method for analyzing large datasets with high-dimensional features per observation, increasing data interpretability while preserving maximum information and enabling visualization of multidimensional data.",A dimensionality reduction method for analyzing large datasets with high-dimensional features per observation increasing data interpretability while preserving maximum information and enabling visualization.,https://en.wikipedia.org/wiki/Principal_component_analysis|,,,,,AIO:DimensionalityReduction -AIO:tDistributedStochasticNeighborembedding,Machine Learning,AIO:MachineLearningSubset,tSNE|t-SNE,,t-Distributed Stochastic Neighbor embedding,,,A dimensionality reduction for visualizing high-dimensional data by giving each datapoint a location in a two or three-dimensional map.,https://en.wikipedia.org/wiki/T-distributed_stochastic_neighbor_embedding|,,,,,AIO:DimensionalityReduction -AIO:ManifoldLearning,Machine Learning,AIO:MachineLearningSubset,,,Manifold Learning,0.25,Methods based on the assumption that observed data lie on a low-dimensional manifold embedded in a higher-dimensional space.,A dimensionality reduction method based on the assumption that observed data lie on a low-dimensional manifold embedded in a higher-dimensional space.,https://arxiv.org/abs/2011.01307|,,,,,AIO:DimensionalityReduction -AIO:RandomForest,Machine Learning,AIO:MachineLearningSubset,,,Random Forest,0.19,"An ensemble learning method for classification, regression, and other tasks that constructs a multitude of decision trees during training.",An ensemble learning method for classification regression and other tasks that constructs a multitude of decision trees during training.,https://en.wikipedia.org/wiki/Random_forest|,,,,,AIO:EnsembleLearning -AIO:ExtremeLearningMachine,Network,AIO:NetworkSubset,ELM,,Extreme Learning Machine,0.76,"Extreme learning machines are feedforward neural networks with randomly assigned hidden node parameters that are not updated, learning output weights in a single step.",A feedback network with randomly assigned hidden nodes that are not updated during training.,https://en.wikipedia.org/wiki/Extreme_Learning_machine,"Input, Hidden, Output","Layers: Input, Hidden, Output",,AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:FeedbackNetwork -AIO:ELUFunction,Function,AIO:FunctionSubset,Exponential Linear Unit|ELU,,ELU Function,0.68,The exponential linear unit (ELU) with alpha > 0 is: x if x > 0 and alpha * (exp(x) - 1) if x < 0 The ELU hyperparameter alpha controls the value to which an ELU saturates for negative net inputs. ELUs diminish the vanishing gradient effect. ELUs have negative values which pushes the mean of the activations closer to zero. Mean activations that are closer to zero enable faster Learning as they bring the gradient closer to the natural gradient. ELUs saturate to a negative value when the argument gets smaller. Saturation means a small derivative which decreases the variation and the information that is propagated to the next layer.,An activation function that is x if x > 0 and alpha * (exp(x) - 1) if x < 0 where alpha controls the value to which an ELU saturates for negative net inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/elu,,,,,AIO:Function -AIO:GELUFunction,Function,AIO:FunctionSubset,Gaussian Error Linear Unit|GELU,,GELU Function,0.58,"Gaussian error linear unit (GELU) computes x * P(X <= x), where P(X) ~ N(0, 1). The (GELU) nonlinearity weights inputs by their value, rather than gates inputs by their sign as in ReLU.",An activation function that computes x * P(X <= x) where P(X) ~ N(0 1) weighting inputs by their value rather than gating inputs by their sign as in ReLU.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/gelu,,,,,AIO:Function -AIO:TanhFunction,Function,AIO:FunctionSubset,hyperbolic tangent,,Tanh Function,0.70,Hyperbolic tangent activation function.,An activation function that is the hyperbolic tangent activation function.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/tanh,,,,,AIO:Function -AIO:ReLUFunction,Function,AIO:FunctionSubset,Rectified Linear Unit|ReLU,,ReLU Function,0.50,"The ReLU activation function returns: max(x, 0), the element-wise maximum of 0 and the input tensor.",An activation function that returns max(x 0) the element-wise maximum of 0 and the input tensor.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/relu,,,,,AIO:Function -AIO:SELUFunction,Function,AIO:FunctionSubset,Scaled Exponential Linear Unit|SELU,,SELU Function,0.16,The SELU activation function multiplies scale (> 1) with the output of the ELU function to ensure a slope larger than one for positive inputs.,An activation function that multiplies scale (> 1) with the output of the ELU function to ensure a slope larger than one for positive inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/selu,,,,,AIO:Function -AIO:SigmoidFunction,Function,AIO:FunctionSubset,,,Sigmoid Function,0.66,"Applies the sigmoid activation function sigmoid(x) = 1 / (1 + exp(-x)). For small values (<-5), sigmoid returns a value close to zero, and for large values (>5) the result of the function gets close to 1. Sigmoid is equivalent to a 2-element Softmax, where the second element is assumed to be zero. The sigmoid function always returns a value between 0 and 1.",An activation function that applies the sigmoid activation function sigmoid(x) = 1 / (1 + exp(-x)) always returning a value between 0 and 1.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/sigmoid,,,,,AIO:Function -AIO:LinearFunction,Function,AIO:FunctionSubset,,,Linear Function,0.36,A linear function has the form f(x) = a + bx.,An activation function that has the form f(x) = a + bx.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/linear,,,,,AIO:Function -AIO:HardSigmoidFunction,Function,AIO:FunctionSubset,,,Hard Sigmoid Function,0.71,A faster approximation of the sigmoid activation. Piecewise linear approximation of the sigmoid function. Ref: 'https://en.wikipedia.org/wiki/Hard_sigmoid',An activation function that is a faster approximation of the sigmoid activation using a piecewise linear approximation.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/hard_sigmoid,,,,,AIO:Function -AIO:SoftplusFunction,Function,AIO:FunctionSubset,,,Softplus Function,0.64,softplus(x) = log(exp(x) + 1),An activation function that is softplus(x) = log(exp(x) + 1).,https://www.tensorflow.org/api_docs/python/tf/keras/activations/softplus,,,,,AIO:Function -AIO:SoftsignFunction,Function,AIO:FunctionSubset,,,Softsign Function,0.54,softsign(x) = x / (abs(x) + 1),An activation function that is softsign(x) = x / (abs(x) + 1).,https://www.tensorflow.org/api_docs/python/tf/keras/activations/softsign,,,,,AIO:Function -AIO:ExponentialFunction,Function,AIO:FunctionSubset,,,Exponential Function,0.47,The exponential function is a mathematical function denoted by f(x)=exp or e^{x}.,An activation function that is the mathematical function denoted by f(x)=exp or e^{x}.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/exponential,,,,,AIO:Function -AIO:SwishFunction,Function,AIO:FunctionSubset,,,Swish Function,0.56,"x*sigmoid(x). It is a smooth, non-monotonic function that consistently matches or outperforms ReLU on deep networks, it is unbounded above and bounded below.",An activation function that is x*sigmoid(x) a smooth non-monotonic function that consistently matches or outperforms ReLU on deep networks.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/swish,,,,,AIO:Function -AIO:SoftmaxFunction,Function,AIO:FunctionSubset,,,Softmax Function,0.70,"The elements of the output vector are in range (0, 1) and sum to 1. Each vector is handled independently. The axis argument sets which axis of the input the function is applied along. Softmax is often used as the activation for the last layer of a classification network because the result could be interpreted as a probability distribution. The softmax of each vector x is computed as exp(x) / tf.reduce_sum(exp(x)). The input values in are the log-odds of the resulting probability.",An activation function where the elements of the output vector are in range (0 1) and sum to 1 and each vector is handled independently.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/softmax,,,,,AIO:Function -AIO:GraphConvolutionalPolicyNetwork,Network,AIO:NetworkSubset,GPCN,,Graph Convolutional Policy Network,0.39,"A graph convolutional policy network (GCPN) generates goal-directed graphs using a graph convolutional network and reinforcement learning, optimizing for domain-specific rewards and adversarial loss",A graph convolutional network that generates goal-directed graphs using reinforcement learning and optimizing for rewards and adversarial loss.,https://arxiv.org/abs/1806.02473,"Input, Hidden, Hidden, Policy, Output","Layers: Input, Hidden, Hidden, Policy, Output",,AIO:InputLayer|AIO:HiddenLayer|AIO:PolicyLayer|AIO:OutputLayer,AIO:GraphConvolutionalNetwork -AIO:GroupthinkBias,Bias,AIO:BiasSubset,Groupthink,,Groupthink Bias,0.00,A psychological phenomenon where people in a group make non-optimal decisions due to a desire to conform or fear of dissent.,A psychological phenomenon where people in a group make non-optimal decisions due to a desire to conform or fear of dissent.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:GroupBias -AIO:FundingBias,Bias,AIO:BiasSubset,,,Funding Bias,0.14,Bias arising when biased results are reported to support or satisfy the funding agency or financial supporter of a research study.,A bias arising when biased results are reported to support or satisfy the funding agency or financial supporter of a research study.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:GroupBias -AIO:DeploymentBias,Bias,AIO:BiasSubset,,,Deployment Bias,0.36,"Arises when systems are used as decision aids for humans, since the human intermediary may act on predictions in ways that are typically not modeled in the system. However, it is still individuals using the deployed system.",A bias arising when systems are used as decision aids for humans since the human intermediary may act on predictions in ways that are typically not modeled in the system.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:GroupBias -AIO:ProbabilisticHiddenLayer,Layer,AIO:LayerSubset,,,Probabilistic Hidden Layer,,,A hidden layer that estimates the probability of a sample being within a certain category.,,,,,,AIO:HiddenLayer -AIO:SpikingHiddenLayer,Layer,AIO:LayerSubset,,,Spiking Hidden Layer,,,"A hidden layer that makes connections to an additional, heterogeneous hidden layer; modeled after biological neural networks.",https://doi.org/10.1016/S0893-6080(97)00011-7,,,,,AIO:HiddenLayer -AIO:GroupBias,Bias,AIO:BiasSubset,In-group Favoritism|In-group–out-group Bias|In-group bias|Intergroup bias|In-group preference,,Group Bias,0.48,"Favoring members of one's in-group over out-group members, expressed in evaluation, resource allocation, and other ways.",A bias characterized by favoring members of one's in-group over out-group members expressed in evaluation resource allocation and other ways.,https://en.wikipedia.org/wiki/In-group_favoritism|GTP-4o with Seppala et al. 2017,,,,,AIO:HumanBias -AIO:RescalingLayer,Layer,AIO:LayerSubset,,,Rescaling Layer,0.17,A preprocessing layer which rescales input values to a new range.,A preprocessing layer that rescales input values to a new range.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Rescaling,,,,,AIO:ImagePreprocessingLayer -AIO:ResizingLayer,Layer,AIO:LayerSubset,,,Resizing Layer,0.86,"A preprocessing layer which resizes images. This layer resizes an image input to a target height and width. The input should be a 4D (batched) or 3D (unbatched) tensor in ""channels_last"" format. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. This layer can be called on tf.RaggedTensor batches of input images of distinct sizes, and will resize the outputs to dense tensors of uniform size.",A preprocessing layer that resizes images to a target size.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Resizing,,,,,AIO:ImagePreprocessingLayer -AIO:CenterCropLayer,Layer,AIO:LayerSubset,,,CenterCrop Layer,0.77,"A preprocessing layer which crops images. This layers crops the central portion of the images to a target size. If an image is smaller than the target size, it will be resized and cropped so as to return the largest possible window in the image that matches the target aspect ratio. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats.",An image preprocessing layer that crops the central portion of images to a target size.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/CenterCrop,,,,,AIO:ImagePreprocessingLayer -AIO:AutomationComplacencyBias,Bias,AIO:BiasSubset,Automation Complaceny,,Automation Complacency Bias,0.73,"Over-reliance on automated systems, leading to attenuated human skills, such as with spelling and autocorrect.",A bias characterized by over-reliance on automated systems leading to attenuated human skills.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:AvailabilityHeuristicBias,Bias,AIO:BiasSubset,Availability Heuristic|Availability Bias,,Availability Heuristic Bias,0.28,A mental shortcut where easily recalled information is overweighted in judgment and decision-making.,A cognitive bias characterized by a mental shortcut where easily recalled information is overweighted in judgment and decision-making.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:ImplicitBias,Bias,AIO:BiasSubset,Confirmatory Bias,,Implicit Bias,0.73,"Unconscious beliefs, attitudes, feelings, associations, or stereotypes that affect information processing, decision-making, and actions.",An individual bias characterized by unconscious beliefs attitudes feelings associations or stereotypes that affect information processing decision-making and actions.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:RashomonEffectBias,Bias,AIO:BiasSubset,Rashomon Effect|Rashomon Principle,,Rashomon Effect Bias,0.54,"Differences in perspective, memory, recall, interpretation, and reporting of the same event by multiple persons or witnesses.",An individual bias characterized by differences in perspective memory recall interpretation and reporting of the same event by multiple persons or witnesses.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:StreetlightEffectBias,Bias,AIO:BiasSubset,Streetlight Effect,,Streetlight Effect Bias,0.31,Bias where people search only where it is easiest to look.,An individual bias where people search only where it is easiest to look.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:ConsumerBias,Bias,AIO:BiasSubset,,,Consumer Bias,0.20,"Bias arising when an algorithm or platform provides users a venue to express their biases, occurring from either side in a digital interaction.",A bias arising when an algorithm or platform provides users a venue to express their biases occurring from either side in a digital interaction.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:ModeConfusionBias,Bias,AIO:BiasSubset,,,Mode Confusion Bias,0.30,"When modal interfaces confuse human operators, causing actions appropriate for a different mode but incorrect for the current situation.",A bias occurring when modal interfaces confuse human operators causing actions appropriate for a different mode but incorrect for the current situation.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:AnchoringBias,Bias,AIO:BiasSubset,,,Anchoring Bias,0.38,"The influence of a reference point or anchor on decisions, leading to insufficient adjustment from that anchor point.",A cognitive bias characterized by the influence of a reference point or anchor on decisions leading to insufficient adjustment from that anchor point.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:ConfirmationBias,Bias,AIO:BiasSubset,,,Confirmation Bias,0.42,"The tendency to prefer information that confirms existing beliefs, influencing the search for, interpretation of, and recall of information.",A cognitive bias characterized by the tendency to prefer information that confirms existing beliefs influencing the search for interpretation of and recall of information.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:CognitiveBias,Bias,AIO:BiasSubset,,,Cognitive Bias,0.29,"Systematic deviation from rational judgment and decision-making, including adaptive mental shortcuts known as heuristics.",A systematic deviation from rational judgment and decision-making including adaptive mental shortcuts known as heuristics.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:PresentationBias,Bias,AIO:BiasSubset,,,Presentation Bias,0.36,"Bias arising from how information is presented on the Web, via a user interface, due to rating or ranking of output, or through users' self-selected, biased interaction.",An individual bias arising from how information is presented on the Web via a user interface due to rating or ranking of output or through users' self-selected biased interaction.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:UserInteractionBias,Bias,AIO:BiasSubset,,,User Interaction Bias,0.45,"Bias arising when a user imposes their own biases during interaction with data, output, results, etc.",An individual bias arising when a user imposes their own biases during interaction with data output results etc.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:BehavioralBias,Bias,AIO:BiasSubset,,,Behavioral Bias,0.45,"Systematic distortions in user behavior across platforms or contexts, or across users represented in different datasets.",An individual bias characterized by systematic distortions in user behavior across platforms or contexts or across users represented in different datasets.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:SelectiveAdherenceBias,Bias,AIO:BiasSubset,,,Selective Adherence Bias,0.37,The tendency to selectively adopt algorithmic advice that matches pre-existing beliefs and stereotypes.,An individual bias characterized by the tendency to selectively adopt algorithmic advice that matches pre-existing beliefs and stereotypes.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:LossOfSituationalAwarenessBias,Bias,AIO:BiasSubset,,,Loss Of Situational Awareness Bias,0.33,"When automation leads to humans being unaware of their situation, making them unprepared to assume control in cooperative systems.",An individual bias occurring when automation leads to humans being unaware of their situation making them unprepared to assume control in cooperative systems.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:AnnotatorReportingBias,Bias,AIO:BiasSubset,,,Annotator Reporting Bias,0.29,When users rely on automation as a heuristic replacement for their own information seeking and processing.,An individual bias occurring when users rely on automation as a heuristic replacement for their own information seeking and processing.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:HumanReportingBias,Bias,AIO:BiasSubset,,,Human Reporting Bias,0.29,When users rely on automation as a heuristic replacement for their own information seeking and processing.,An individual bias that arises when users depend on automated systems as heuristic substitutes for their own information-seeking and processing efforts.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:InterpretationBias,Bias,AIO:BiasSubset,,,Interpretation Bias,0.35,A form of information processing bias where users interpret algorithmic outputs according to their internalized biases and views.,An individual bias where users interpret algorithmic outputs according to their internalized biases and views.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:IndividualBias -AIO:NoisyInputLayer,Layer,AIO:LayerSubset,,,Noisy Input Layer,,,An input layer that adds noise to each value.,https://doi.org/10.1109/21.155944,,,,,AIO:InputLayer -AIO:BackfedInputLayer,Layer,AIO:LayerSubset,,,Backfed Input Layer,,,An input layer that receives values from another layer.,,,,,,AIO:InputLayer -AIO:MatchedInputOutputLayer,Layer,AIO:LayerSubset,,,Matched Input-Output Layer,,,An input layer with a shape corresponding to that of the output layer.,,,,,,AIO:InputLayer -AIO:MaskedLanguageModel,Model,AIO:ModelSubset,Masked Language Model,bidirectional encoder|denoising autoencoder,Masked Language Model,0.16,"A masked language model is trained to predict randomly masked tokens in a sequence, based on the remaining unmasked tokens. This allows it to build deep bidirectional representations that can be effectively transferred to various NLP tasks via fine-tuning.",A language model that is trained to predict randomly masked tokens in a sequence based on the remaining unmasked tokens allowing it to build deep bidirectional representations that can be effectively transferred to various NLP tasks via fine-tuning.,,,,,,AIO:LanguageModel -AIO:ImplicitLanguageModel,Model,AIO:ModelSubset,Implicit LM,Energy-based models|Token-level scoring,Implicit Language Model,0.22,"An implicit language model uses an energy function to score entire sequences instead of factorizing probabilities autoregressively, better capturing global properties and long-range dependencies.",A language model that uses an energy function to score entire sequences instead of factorizing probabilities autoregressively better capturing global properties and long-range dependencies.,https://arxiv.org/pdf/2303.16189|,,,,,AIO:LanguageModel -AIO:AutoregressiveLanguageModel,Model,AIO:ModelSubset,Autoregressive Language Model,generative language model|sequence-to-sequence model,Autoregressive Language Model,0.28,"An autoregressive language model generates text sequentially, predicting one token at a time based on the previously generated tokens. It excels at natural language generation tasks by modeling the probability distribution over sequences of tokens.",A language model that generates text sequentially predicting one token at a time based on the previously generated tokens excelling at natural language generation tasks by modeling the probability distribution over sequences of tokens.,,,,,,AIO:LanguageModel -AIO:GenerativeLanguageInterface,Model,AIO:ModelSubset,Generative Language Interface,Interactive generation,Generative Language Interface,0.23,"A generative language interface enables users to engage in an interactive dialogue with an LLM, providing feedback to guide and refine the generated outputs iteratively.",A language model that enables users to engage in an interactive dialogue with an LLM providing feedback to guide and refine the generated outputs iteratively.,,,,,,AIO:LanguageModel -AIO:HierarchicalLanguageModel,Model,AIO:ModelSubset,Hierarchical LM,multi-scale representations,Hierarchical Language Model,0.10,"A hierarchical language model represents language at multiple levels of granularity, learning hierarchical representations that capture both low-level patterns and high-level abstractions.",A language model that represents language at multiple levels of granularity learning hierarchical representations that capture both low-level patterns and high-level abstractions.,https://doi.org/10.1016/j.ipm.2024.103698|,,,,,AIO:LanguageModel -AIO:RecursiveLLM,Model,AIO:ModelSubset,Recursive Large Language Model|Self-Attending Large Language Model,self-attention|iterative refinement,Recursive LLM,0.18,"A recursive language model uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions, improving systematic generalization abilities.",A large language model that uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions improving systematic generalization abilities.,https://doi.org/10.1609/aaai.v33i01.33017450|,,,,,AIO:LanguageModel -AIO:GraphLanguageModel,Model,AIO:ModelSubset,Graph LM,Structured representations,Graph Language Model,0.17,"A graph language model operates over structured inputs or outputs represented as graphs, enabling reasoning over explicit relational knowledge representations during language tasks.",A language model that operates over structured inputs or outputs represented as graphs enabling reasoning over explicit relational knowledge representations during language tasks.,https://arxiv.org/abs/2401.07105|,,,,,AIO:LanguageModel -AIO:FactoredLanguageModel,Model,AIO:ModelSubset,Factorized Language Model,,Factored Language Model,0.38,"A factored language model views each word as a vector of multiple factors, such as part-of-speech, morphology, and semantics, to improve language modeling.",A language model that views each word as a vector of multiple factors such as part-of-speech morphology and semantics to improve language modeling.,https://en.wikipedia.org/wiki/Factored_language_model|,,,,,AIO:LanguageModel -AIO:LargeLanguageModel,Model,AIO:ModelSubset,LLM,,Large Language Model,0.07,"A language model consisting of a neural network with many parameters (typically billions of weights or more), trained on large quantities of unlabeled text using self-supervised learning or semi-supervised learning.",A language model consisting of a neural network with many parameters (typically billions of weights or more) trained on large quantities of unlabeled text using self-supervised learning or semi-supervised learning.,https://en.wikipedia.org/wiki/Large_language_model|,,,,,AIO:LanguageModel -AIO:ModularLanguageModel,Model,AIO:ModelSubset,Modular LM,,Modular Language Model,0.07,"A modular language model consists of multiple specialized components or skills that can be dynamically composed and recombined to solve complex tasks, mimicking the modular structure of human cognition.",A language model that consists of multiple specialized components or skills that can be dynamically composed and recombined to solve complex tasks mimicking the modular structure of human cognition.,https://arxiv.org/abs/2302.11529v2|,,,,,AIO:LanguageModel -AIO:MultimodalLanguageModel,Model,AIO:ModelSubset,Mulimodal LM,,Multimodal Language Model,0.22,"A multimodal language model learns joint representations across different modalities like text, vision, and audio in an end-to-end fashion for better cross-modal understanding and generation.",A language model that learns joint representations across different modalities like text vision and audio in an end-to-end fashion for better cross-modal understanding and generation.,https://arxiv.org/abs/2205.12630|,,,,,AIO:LanguageModel -AIO:TransformerLanguageModel,Model,AIO:ModelSubset,Transformer LM,,Transformer Language Model,0.19,"A transformer LM is a neural network model that uses the transformer architecture based on multi-head attention mechanisms, allowing it to contextualize tokens within a context window for effective language understanding and generation.",A language model that uses the transformer architecture based on multi-head attention mechanisms allowing it to contextualize tokens within a context window for effective language understanding and generation.,https://arxiv.org/abs/1706.03762 | |,,,,,AIO:LanguageModel -AIO:LifelongLearningLLM,Model,AIO:ModelSubset,Lifelong Learning LLM|Continual Learning LLM|Forever Learning,Catastrophic forgetting|Plasticity-Stability balance,Lifelong Learning LLM,0.33,"A lifelong learning LLM continually acquires new knowledge over time without forgetting previously learned information, maintaining a balance between plasticity and stability.",A large language model that continually acquires new knowledge over time without forgetting previously learned information maintaining a balance between plasticity and stability.,,,,,,AIO:LargeLanguageModel -AIO:MixtureofExpertsLLM,Model,AIO:ModelSubset,MoE Large Language Model|Mixture-of-Experts Large Language Model,conditional computation|model parallelism,Mixture-of-Experts LLM,,,"A LLM which dynamically selects and combines outputs from multiple expert submodels, allowing for efficient scaling by conditionally activating only a subset of model components for each input.",https://proceedings.mlr.press/v162/du22c.html|,,,,,AIO:LargeLanguageModel -AIO:ControllableLLM,Model,AIO:ModelSubset,Controllable Large Language Model,conditional generation|guided generation,Controllable LLM,0.43,"A controllable LLM allows for explicit control over certain attributes of the generated text, such as style, tone, topic, or other desired characteristics, through conditioning or specialized training objectives.",A large language model that allows for explicit control over certain attributes of the generated text such as style tone topic or other desired characteristics through conditioning or specialized training objectives.,,,,,,AIO:LargeLanguageModel -AIO:FactorizedLLM,Model,AIO:ModelSubset,Factorized Large Language Model|Factorized Learning Assisted with Large Language Model,Conditional masking|Product of experts,Factorized LLM,0.21,"A factorized LLM decomposes the full language modeling task into multiple sub-components or experts that each focus on a subset of the information, enabling more efficient scaling.",A large language model that decomposes the full language modeling task into multiple sub-components or experts that each focus on a subset of the information enabling more efficient scaling.,https://doi.org/10.48550/arXiv.2403.12556|,,,,,AIO:LargeLanguageModel -AIO:DialogueLLM,Model,AIO:ModelSubset,Dialogue Large Language Model,conversational AI|multi-turn dialogue,Dialogue LLM,0.39,"A dialogue LLM is optimized for engaging in multi-turn conversations, understanding context, and generating relevant, coherent responses continuously over many dialogue turns.",A large language model that is optimized for engaging in multi-turn conversations understanding context and generating relevant coherent responses continuously over many dialogue turns.,,,,,,AIO:LargeLanguageModel -AIO:MultilingualLLM,Model,AIO:ModelSubset,Multilingual Large Language Model,cross-lingual transfer,Multilingual LLM,0.27,"A multilingual LLM is trained on text from multiple languages, learning shared representations that enable zero-shot or few-shot transfer to new languages.",A large language model that is trained on text from multiple languages learning shared representations that enable zero-shot or few-shot transfer to new languages.,,,,,,AIO:LargeLanguageModel -AIO:MultimodalFusionLLM,Model,AIO:ModelSubset,Multimodal Fusion LLM,cross-modal grounding,Multimodal Fusion LLM,0.37,"A multimodal fusion LLM learns joint representations across different modalities like text, vision, and audio in an end-to-end fashion for better cross-modal understanding and generation.",A large language model that learns joint representations across different modalities like text vision and audio in an end-to-end fashion for better cross-modal understanding and generation.,,,,,,AIO:LargeLanguageModel -AIO:CrossDomainLLM,Model,AIO:ModelSubset,Domain-General LLM,domain adaptation|cross-domain transfer,Cross-Domain LLM,,,"A LLM that performs well across a wide range of domains without significant loss in performance, facilitated by advanced domain adaptation techniques.",,,,,,AIO:LargeLanguageModel -AIO:DifferentiableLLM,Model,AIO:ModelSubset,Differentiable Large Language Model,end-to-end training|fully backpropagable,Differentiable LLM,0.31,"A differentiable LLM has an architecture amenable to full end-to-end training via backpropagation, without relying on teacher forcing or unlikelihood training objectives.",A large language model that has an architecture amenable to full end-to-end training via backpropagation without relying on teacher forcing or unlikelihood training objectives.,,,,,,AIO:LargeLanguageModel -AIO:EnergyBasedLLM,Model,AIO:ModelSubset,Energy-Based Large Language Model,energy scoring|explicit density modeling,Energy-Based LLM,,,"A LLM which models the explicit probability density over token sequences using an energy function, rather than an autoregressive factorization. This can improve modeling of long-range dependencies and global coherence.",,,,,,AIO:LargeLanguageModel -AIO:EvolutionaryLLM,Model,AIO:ModelSubset,Evolutionary Language Model,evolutionary algorithms|genetic programming,Evolutionary LLM,0.38,"An evolutionary LLM applies principles of evolutionary computation to optimize its structure and parameters, evolving over time to improve performance.",A large language model that applies principles of evolutionary computation to optimize its structure and parameters evolving over time to improve performance.,,,,,,AIO:LargeLanguageModel -AIO:MemoryAugmentedLLM,Model,AIO:ModelSubset,Memory-Augmented Large Language Model,external memory,Memory-Augmented LLM,,,"A LLM which incorporates external writable and readable memory components, allowing it to store and retrieve information over long contexts.",https://arxiv.org/abs/2306.07174|,,,,,AIO:LargeLanguageModel -AIO:KnowledgeGroundedLLM,Model,AIO:ModelSubset,Knowledge-Grounded Large Language Model,factual grounding|knowledge integration,Knowledge-Grounded LLM,,,"A LLM which incorporates external knowledge sources or knowledge bases into the model architecture, enabling it to generate more factually accurate and knowledge-aware text.",,,,,,AIO:LargeLanguageModel -AIO:MetaLearningLLM,Model,AIO:ModelSubset,Meta-Learning Large Language Model,few-shot adaptation|learning to learn,Meta-Learning LLM,,,"A LLM which is trained in a way that allows it to quickly adapt to new tasks or datasets through only a few examples or fine-tuning steps, leveraging meta-learned priors about how to efficiently learn.",,,,,,AIO:LargeLanguageModel -AIO:PromptbasedFineTuningLLM,Model,AIO:ModelSubset,Prompt-tuned Large Language Model|Prompt-based Fine-Tuning Large Language Model,in-context learning|few-shot learning,Prompt-based Fine-Tuning LLM,,,"A LLM which is fine-tuned on a small number of examples or prompts, rather than full task datasets. This allows for rapid adaptation to new tasks with limited data, leveraging the model's few-shot learning capabilities.",,,,,,AIO:LargeLanguageModel -AIO:LanguageInterfaceLLM,Model,AIO:ModelSubset,Language Interface LLM,Interactive learning,Language Interface LLM,0.29,"A language interface LLM supports interactive semantic parsing, enabling users to provide feedback and corrections to dynamically refine and update the language model.",A large language model that supports interactive semantic parsing enabling users to provide feedback and corrections to dynamically refine and update the language model.,,,,,,AIO:LargeLanguageModel -AIO:ExplainableLLM,Model,AIO:ModelSubset,Explainable Language Model|XAI LLM,interpretability|model understanding,Explainable LLM,0.39,"An explainable LLM is designed to provide insights into its decision-making process, making it easier for users to understand and trust the model's outputs. It incorporates mechanisms for interpreting and explaining its predictions in human-understandable terms.",A large language model that is designed to provide insights into its decision-making process making it easier for users to understand and trust the model's outputs by incorporating mechanisms for interpreting and explaining its predictions in human-understandable terms.,,,,,,AIO:LargeLanguageModel -AIO:RetrievalAugmentedLLM,Model,AIO:ModelSubset,Retrieval-Augmented Large Language Model,knowledge grounding|open-book question answering,Retrieval-Augmented LLM,,,"A LLM which combines a pre-trained language model with a retrieval system that can access external knowledge sources. This allows the model to condition its generation on relevant retrieved knowledge, improving factual accuracy and knowledge grounding.",,,,,,AIO:LargeLanguageModel -AIO:NeuroSymbolicLLM,Model,AIO:ModelSubset,Neuro-Symbolic Large Language Model,knowledge reasoning|symbolic grounding,Neuro-Symbolic LLM,,,"A LLM which combines neural language modeling with symbolic reasoning components, leveraging structured knowledge representations and logical inferences to improve reasoning capabilities.",,,,,,AIO:LargeLanguageModel -AIO:CurriculumLearningLLM,Model,AIO:ModelSubset,Curriculum Learning LLM,Learning progression,Curriculum Learning LLM,0.30,"A curriculum learning LLM is trained by presenting learning examples in a meaningful order from simple to complex, mimicking the learning trajectory followed by humans.",A large language model that is trained by presenting learning examples in a meaningful order from simple to complex mimicking the learning trajectory followed by humans.,,,,,,AIO:LargeLanguageModel -AIO:ContinualLearningLLM,Model,AIO:ModelSubset,CL-Large Language Model|Continual Learning Large Language Model,lifelong learning|catastrophic forgetting,Continual Learning LLM,0.31,A continual learning LLM continually acquires new knowledge and skills over time without forgetting previously learned information. This allows the model to adapt and expand its capabilities as new data becomes available.,A large language model that continually acquires new knowledge and skills over time without forgetting previously learned information allowing the model to adapt and expand its capabilities as new data becomes available.,,,,,,AIO:LargeLanguageModel -AIO:SparseLLM,Model,AIO:ModelSubset,Sparse Large Language Model,model compression|parameter efficiency,Sparse LLM,0.22,"A sparse LLM uses techniques like pruning or quantization to reduce the number of non-zero parameters in the model, making it more parameter-efficient and easier to deploy on resource-constrained devices.",A large language model that uses techniques like pruning or quantization to reduce the number of non-zero parameters in the model making it more parameter-efficient and easier to deploy on resource-constrained devices.,,,,,,AIO:LargeLanguageModel -AIO:EmbodiedLLM,Model,AIO:ModelSubset,Embodied Large Language Model,multimodal grounding,Embodied LLM,0.42,"An embodied LLM integrates language with other modalities like vision, audio, and robotics to enable grounded language understanding in real-world environments.",A large language model that integrates language with other modalities like vision audio and robotics to enable grounded language understanding in real-world environments.,,,,,,AIO:LargeLanguageModel -AIO:InstructionTunedLLM,Model,AIO:ModelSubset,Instruction-Tuned Large Language Model,natural language instructions|constitutional AI,Instruction-Tuned LLM,,,"A LLM which is fine-tuned to follow natural language instructions accurately and safely, learning to map from instructions to desired model behavior in a more controlled and principled way.",,,,,,AIO:LargeLanguageModel -AIO:GenerativeCommonsenseLLM,Model,AIO:ModelSubset,Generative Commonsense Large Language Model|World Model,physical reasoning|causal modeling,Generative Commonsense LLM,0.38,"A generative commonsense LLM is trained to understand and model basic physics, causality, and common sense about how the real world works.",A large language model that is trained to understand and model basic physics causality and common sense about how the real world works.,https://arxiv.org/abs/2306.12672|,,,,,AIO:LargeLanguageModel -AIO:SelfSupervisedLLM,Model,AIO:ModelSubset,Self-Supervised LLM,Pretext tasks,Self-Supervised LLM,,,"A LLM which learns rich representations by solving pretext tasks that involve predicting parts of the input from other observed parts of the data, without relying on human-annotated labels.",,,,,,AIO:LargeLanguageModel -AIO:FederatedLLM,Model,AIO:ModelSubset,Federated Large Language Model,privacy-preserving|decentralized training,Federated LLM,0.34,"A federated LLM is trained in a decentralized manner across multiple devices or silos, without directly sharing private data. This enables collaborative training while preserving data privacy and security.",A large language model that is trained in a decentralized manner across multiple devices or silos without directly sharing private data enabling collaborative training while preserving data privacy and security.,,,,,,AIO:LargeLanguageModel -AIO:OrdinalLLM,Model,AIO:ModelSubset,Ordinal Large Language Model,ranking|preference modeling,Ordinal LLM,0.33,"An ordinal LLM is trained to model ordinal relationships and rank outputs, rather than model probability distributions over text sequences directly.",A large language model that is trained to model ordinal relationships and rank outputs rather than model probability distributions over text sequences directly.,,,,,,AIO:LargeLanguageModel -AIO:ReasoningLLM,Model,AIO:ModelSubset,Reasoning Large Language Model|Rational Large Language Model,reasoning|logical inferences,Reasoning LLM,0.38,"A reasoning LLM incorporates explicit reasoning capabilities, leveraging logical rules, axioms, or external knowledge to make deductive inferences during language tasks.",A large language model that incorporates explicit reasoning capabilities leveraging logical rules axioms or external knowledge to make deductive inferences during language tasks.,https://doi.org/10.18653/v1/2023.acl-long.347|,,,,,AIO:LargeLanguageModel -AIO:ContrastiveLearningLLM,Model,AIO:ModelSubset,Contrastive Learning LLM,Representation learning,Contrastive Learning LLM,0.25,"A contrastive learning LLM is trained to pull semantically similar samples closer together and push dissimilar samples apart in the representation space, learning high-quality features useful for downstream tasks.",A large language model that is trained to pull semantically similar samples closer together and push dissimilar samples apart in the representation space learning high-quality features useful for downstream tasks.,,,,,,AIO:LargeLanguageModel -AIO:LowResourceLLM,Model,AIO:ModelSubset,Low-Resource Language Model,resource-efficient|low-resource languages,Low-Resource LLM,,,"A LLM which is optimized for performance in scenarios with limited data, computational resources, or for languages with sparse datasets.",,,,,,AIO:LargeLanguageModel -AIO:ReinforcementLearningLLM,Model,AIO:ModelSubset,RL-Large Language Model|Reinforcement Learning Large Language Model,reward modeling|decision transformers,Reinforcement Learning LLM,0.43,"An RL-LLM is a language model fine-tuned using reinforcement learning, where the model receives rewards for generating text that satisfies certain desired properties or objectives. This can improve the quality, safety, or alignment of generated text.",A large language model that is fine-tuned using reinforcement learning where the model receives rewards for generating text that satisfies certain desired properties or objectives improving the quality safety or alignment of generated text.,,,,,,AIO:LargeLanguageModel -AIO:UnsupervisedLLM,Model,AIO:ModelSubset,Unsupervised Large Language Model,self-supervised,Unsupervised LLM,0.36,"An unsupervised LLM is trained solely on unlabeled data using self-supervised objectives like masked language modeling, without any supervised fine-tuning.",A large language model that is trained solely on unlabeled data using self-supervised objectives like masked language modeling without any supervised fine-tuning.,,,,,,AIO:LargeLanguageModel -AIO:SemiSupervisedLLM,Model,AIO:ModelSubset,Semi-Supervised Large Language Model,self-training,Semi-Supervised LLM,,,A LLM which combines self-supervised pretraining on unlabeled data with supervised fine-tuning on labeled task data.,,,,,,AIO:LargeLanguageModel -AIO:CompositionalGeneralizationLLM,Model,AIO:ModelSubset,Compositional Generalization Large Language Model,systematic generalization|out-of-distribution generalization,Compositional Generalization LLM,0.24,"A compositional generalization LLM is trained to understand and recombine the underlying compositional structures in language, enabling better generalization to novel combinations and out-of-distribution examples.",A large language model that is trained to understand and recombine the underlying compositional structures in language enabling better generalization to novel combinations and out-of-distribution examples.,,,,,,AIO:LargeLanguageModel -AIO:GenerativeAdversarialNetworkAugmentedLLM,Model,AIO:ModelSubset,GAN-Large Language Model|Generative Adversarial Network-Augmented Large Language Model,text generation|adversarial training,Generative Adversarial Network-Augmented LLM,,,"A LLM which incorporates a generative adversarial network (GAN) into its training process, using a discriminator network to provide a signal for generating more realistic and coherent text. This adversarial training can improve the quality and diversity of generated text.",,,,,,AIO:LargeLanguageModel -AIO:MultiTaskLLM,Model,AIO:ModelSubset,Multi-Task Large Language Model,transfer learning,Multi-Task LLM,,,"A LLM which is trained jointly on multiple language tasks simultaneously, learning shared representations that transfer across tasks.",,,,,,AIO:LargeLanguageModel -AIO:TransferLearningLLM,Model,AIO:ModelSubset,Transfer LLM,transfer learning,Transfer Learning LLM,0.28,"A transfer learning LLM leverages knowledge acquired during training on one task to improve performance on different but related tasks, facilitating more efficient learning and adaptation.",A large language model that leverages knowledge acquired during training on one task to improve performance on different but related tasks facilitating more efficient learning and adaptation.,,,,,,AIO:LargeLanguageModel -AIO:DomainAdaptedLLM,Model,AIO:ModelSubset,Domain-Adapted Large Language Model,transfer learning|domain robustness,Domain-Adapted LLM,,,"A LLM which is pre-trained on a broad corpus and then fine-tuned on domain-specific data to specialize its capabilities for particular domains or applications, like scientific literature or code generation.",,,,,,AIO:LargeLanguageModel -AIO:CausalLLM,Model,AIO:ModelSubset,Causal Large Language Model,unidirectional|autoregressive,Causal LLM,0.30,"A causal LLM only attends to previous tokens in the sequence when generating text, modeling the probability distribution autoregressively from left-to-right or causally.",A large language model that only attends to previous tokens in the sequence when generating text modeling the probability distribution autoregressively from left-to-right or causally.,,,,,,AIO:LargeLanguageModel -AIO:PersonalizedLLM,Model,AIO:ModelSubset,Personalized Large Language Model,user adaptation LLM,Personalized LLM,0.38,"A personalized LLM adapts its language modeling and generation to the preferences, style, and persona of individual users or audiences.",A large language model that adapts its language modeling and generation to the preferences style and persona of individual users or audiences.,,,,,,AIO:LargeLanguageModel -AIO:EthicalLLM,Model,AIO:ModelSubset,Ethical Large Language Model,value alignment|constituitional AI,Ethical LLM,0.38,"An ethical LLM is trained to uphold certain ethical principles, values, or rules in its language generation to increase safety and trustworthiness.",A large language model that is trained to uphold certain ethical principles values or rules in its language generation to increase safety and trustworthiness.,,,,,,AIO:LargeLanguageModel -AIO:ZeroShotLearningLLM,Model,AIO:ModelSubset,Zero-Shot LLM,zero-shot learning,Zero-Shot Learning LLM,,,"A LLM which performs tasks or understands concepts it has not explicitly been trained on, demonstrating a high degree of generalization and understanding.",,,,,,AIO:LargeLanguageModel -AIO:DatatoTextLLM,Model,AIO:ModelSubset,Data-to-Text LLM|Meaning representation,,Data-to-Text LLM,,,"A LLM that generates natural language descriptions from structured data sources like tables, graphs, and knowledge bases, requiring grounding in meaning representations.",,,,,,AIO:LargeLanguageModel -AIO:DecoderLLM,Model,AIO:ModelSubset,,,Decoder LLM,0.62,"A decoder-only architecture consisting of only a decoder, trained to predict the next token in a sequence given the previous tokens. Unlike the encoder-decoder architecture, it does not have an explicit encoder and encodes information implicitly in the hidden state of the decoder, updated at each step of the generation process.",A large language model that uses a decoder-only architecture consisting of only a decoder trained to predict the next token in a sequence given the previous tokens.,"https://www.practicalai.io/understanding-transformer-model-architectures/#:~:text=Encoder| -",,,,,AIO:LargeLanguageModel -AIO:EncoderLLM,Model,AIO:ModelSubset,,,Encoder LLM,0.48,"An encoder-only architecture that encodes the input sequence into a fixed-length representation, which is then used as input to a classifier or regressor for prediction. The model has a pre-trained general-purpose encoder that requires fine-tuning for specific tasks.",A large language model that uses an encoder-only architecture to encode the input sequence into a fixed-length representation which is then used as input to a classifier or regressor for prediction.,"https://www.practicalai.io/understanding-transformer-model-architectures/#:~:text=Encoder| -",,,,,AIO:LargeLanguageModel -AIO:EncoderDecoderLLM,Model,AIO:ModelSubset,,,Encoder-Decoder LLM,,,"The LLM introduced in the ""Attention Is All You Need"" paper. The encoder processes the input sequence to generate a hidden representation summarizing the input information, while the decoder uses this hidden representation to generate the desired output sequence.","https://www.practicalai.io/understanding-transformer-model-architectures/#:~:text=Encoder| - -",,,,,AIO:LargeLanguageModel -AIO:ReshapingLayer,Layer,AIO:LayerSubset,Reshape Layer,,Reshaping Layer,0.50,Reshape layers are used to change the shape of the input.,A layer that is used to change the shape of the input.,https://keras.io/api/layers/reshaping_layers/reshape/,,,,,AIO:Layer -AIO:InputLayer,Layer,AIO:LayerSubset,,,Input Layer,0.52,"The input layer of a neural network is composed of artificial input neurons, and brings the initial data into the system for further processing by subsequent layers of artificial neurons. The input layer is the very beginning of the workflow for the artificial neural network.",A layer composed of artificial input neurons that brings the initial data into the system for further processing by subsequent layers.,"https://www.techopedia.com/definition/33262/input-layer-neural-networks#:~:text=Explains%20Input%20Layer-,What%20Does%20Input%20Layer%20Mean%3F,for%20the%20artificial%20neural%20network.",,,,,AIO:Layer -AIO:RecurrentLayer,Layer,AIO:LayerSubset,,,Recurrent Layer,0.37,"A layer of an RNB, composed of recurrent units and with the number of which is the hidden size of the layer.",A layer composed of recurrent units with the number equal to the hidden size of the layer.,https://docs.nvidia.com/deepLearning/performance/dl-performance-recurrent/index.html#recurrent-layer,,,,,AIO:Layer -AIO:OutputLayer,Layer,AIO:LayerSubset,,,Output Layer,0.71,"The output layer in an artificial neural network is the last layer of neurons that produces given outputs for the program. Though they are made much like other artificial neurons in the neural network, output layer neurons may be built or observed in a different way, given that they are the last “actor” nodes on the network.",A layer containing the last neurons in the network that produces given outputs for the program.,https://www.techopedia.com/definition/33263/output-layer-neural-networks,,,,,AIO:Layer -AIO:HiddenLayer,Layer,AIO:LayerSubset,,,Hidden Layer,0.69,"A hidden layer is located between the input and output of the algorithm, in which the function applies weights to the inputs and directs them through an activation function as the output. In short, the hidden layers perform nonlinear transformations of the inputs entered into the network. Hidden layers vary depending on the function of the neural network, and similarly, the layers may vary depending on their associated weights.",A layer located between the input and output that performs nonlinear transformations of the inputs entered into the network.,https://deepai.org/machine-Learning-glossary-and-terms/hidden-layer-machine-Learning,,,,,AIO:Layer -AIO:MemoryCellLayer,Layer,AIO:LayerSubset,,,Memory Cell Layer,,,"A layer of cells, each with an internal state or weights.",https://doi.org/10.1162/neco.1997.9.8.1735,,,,,AIO:Layer -AIO:WeightedLayer,Layer,AIO:LayerSubset,,,Weighted Layer,,,A layer of values to be applied to other cells or neurons in a network.,,,,,,AIO:Layer -AIO:AdditionLayer,Layer,AIO:LayerSubset,,,Addition Layer,,,A layer that adds inputs from one or more other layers to cells or neurons of a target layer.,,,,,,AIO:Layer -AIO:StackedRNNCellsLayer,Layer,AIO:LayerSubset,,,StackedRNNCells Layer,0.52,Wrapper allowing a stack of RNN cells to behave as a single cell. Used to implement efficient stacked RNNs.,A layer that allows a stack of RNN cells to behave as a single cell.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/StackedRNNCells,,,,,AIO:Layer -AIO:ActivationLayer,Layer,AIO:LayerSubset,,,Activation Layer,0.50,Applies an activation function to an output.,A layer that applies an activation function to an output.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Activation,,,,,AIO:Layer -AIO:RegularizationLayer,Layer,AIO:LayerSubset,,,Regularization Layer,0.52,Regularizers allow you to apply penalties on layer parameters or layer activity during optimization. These penalties are summed into the loss function that the network optimizes. Regularization penalties are applied on a per-layer basis.,A layer that applies penalties on layer parameters or layer activity during optimization summed into the loss function that the network optimizes.,https://keras.io/api/layers/regularizers/,,,,,AIO:Layer -AIO:DotLayer,Layer,AIO:LayerSubset,,,Dot Layer,0.73,"Layer that computes a dot product between samples in two tensors. E.g. if applied to a list of two tensors a and b of shape (batch_size, n), the output will be a tensor of shape (batch_size, 1) where each entry i will be the dot product between a[i] and b[i].",A layer that computes a dot product between samples in two tensors.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Dot,,,,,AIO:Layer -AIO:ConvolutionalLayer,Layer,AIO:LayerSubset,,,Convolutional Layer,0.61,"A convolutional layer is the main building block of a CNN. It contains a set of filters (or kernels), parameters of which are to be learned throughout the training. The size of the filters is usually smaller than the actual image. Each filter convolves with the image and creates an activation map.",A layer that contains a set of filters (or kernels) parameters of which are to be learned throughout the training.,"https://www.sciencedirect.com/topics/engineering/convolutional-layer#:~:text=A%20convolutional%20layer%20is%20the,and%20creates%20an%20activation%20map.",,,,,AIO:Layer -AIO:Cropping3DLayer,Layer,AIO:LayerSubset,,,Cropping3D Layer,0.87,Cropping layer for 3D data (e.g. spatial or spatio-temporal).,A layer that crops along spatial dimensions (depth,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Cropping3D,,,,,AIO:Layer -AIO:Cropping2DLayer,Layer,AIO:LayerSubset,,,Cropping2D Layer,0.64,"Cropping layer for 2D input (e.g. picture). It crops along spatial dimensions, i.e. height and width.",A layer that crops along spatial dimensions (i.e. height and width) for 2D input.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Cropping2D,,,,,AIO:Layer -AIO:AttentionLayer,Layer,AIO:LayerSubset,,,Attention Layer,0.79,"Dot-product attention layer, a.k.a. Luong-style attention.",A layer that implements dot-product attention also known as Luong-style attention.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Attention,,,,,AIO:Layer -AIO:NoiseDenseLayer,Layer,AIO:LayerSubset,,,Noise Dense Layer,0.90,"Noisy dense layer that injects random noise to the weights of dense layer. Noisy dense layers are fully connected layers whose weights and biases are augmented by factorised Gaussian noise. The factorised Gaussian noise is controlled through gradient descent by a second weights layer. A NoisyDense layer implements the operation: $$ \mathrm{NoisyDense}(x) = \mathrm{activation}(\mathrm{dot}(x, \mu + (\sigma \cdot \epsilon)) \mathrm{bias}) $$ where mu is the standard weights layer, epsilon is the factorised Gaussian noise, and delta is a second weights layer which controls epsilon.",A layer that is a densely-connected neural network layer with added noise for regularization.,https://www.tensorflow.org/addons/api_docs/python/tfa/layers/NoisyDense,,,,,AIO:Layer -AIO:DenseLayer,Layer,AIO:LayerSubset,,,Dense Layer,0.77,Just your regular densely-connected NN layer.,A layer that is a regular densely-connected neural network layer.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Dense,,,,,AIO:Layer -AIO:MaskingLayer,Layer,AIO:LayerSubset,,,Masking Layer,0.80,"Masks a sequence by using a mask value to skip timesteps. For each timestep in the input tensor (dimension #1 in the tensor), if all values in the input tensor at that timestep are equal to mask_value, then the timestep will be masked (skipped) in all downstream layers (as long as they support masking). If any downstream layer does not support masking yet receives such an input mask, an exception will be raised.",A layer that masks a sequence by using a mask value to skip timesteps.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Masking,,,,,AIO:Layer -AIO:KernelLayer,Layer,AIO:LayerSubset,,,Kernel Layer,,,A layer that obtains the dot product of input values or subsets of input values.,,,,,,AIO:Layer -AIO:CategoricalFeaturesPreprocessingLayer,Layer,AIO:LayerSubset,,,Categorical Features Preprocessing Layer,0.00,A layer that performs categorical data preprocessing operations.,A layer that performs categorical data preprocessing operations.,https://keras.io/guides/preprocessing_layers/,,,,,AIO:Layer -AIO:PreprocessingLayer,Layer,AIO:LayerSubset,,,Preprocessing Layer,0.00,A layer that performs data preprocessing operations.,A layer that performs data preprocessing operations.,https://www.tensorflow.org/guide/keras/preprocessing_layers,,,,,AIO:Layer -AIO:ImageAugmentationLayer,Layer,AIO:LayerSubset,,,Image Augmentation Layer,0.00,A layer that performs image data preprocessing augmentations.,A layer that performs image data preprocessing augmentations.,https://keras.io/guides/preprocessing_layers/,,,,,AIO:Layer -AIO:ImagePreprocessingLayer,Layer,AIO:LayerSubset,,,Image Preprocessing Layer,0.00,A layer that performs image data preprocessing operations.,A layer that performs image data preprocessing operations.,https://keras.io/guides/preprocessing_layers/,,,,,AIO:Layer -AIO:NumericalFeaturesPreprocessingLayer,Layer,AIO:LayerSubset,,,Numerical Features Preprocessing Layer,0.00,A layer that performs numerical data preprocessing operations.,A layer that performs numerical data preprocessing operations.,https://keras.io/guides/preprocessing_layers/,,,,,AIO:Layer -AIO:TextPreprocessingLayer,Layer,AIO:LayerSubset,,,Text Preprocessing Layer,0.00,A layer that performs text data preprocessing operations.,A layer that performs text data preprocessing operations.,https://keras.io/guides/preprocessing_layers/,,,,,AIO:Layer -AIO:GRUCellLayer,Layer,AIO:LayerSubset,,,GRUCell Layer,0.52,"Cell class for the GRU layer. This class processes one step within the whole time sequence input, whereas tf.keras.layer.GRU processes the whole sequence.",A layer that processes one step within the whole time sequence input for a GRU layer.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GRUCell,,,,,AIO:Layer -AIO:SimpleRNNCellLayer,Layer,AIO:LayerSubset,,,SimpleRNNCell Layer,0.63,"Cell class for SimpleRNN. This class processes one step within the whole time sequence input, whereas tf.keras.layer.SimpleRNN processes the whole sequence.",A layer that processes one step within the whole time sequence input for a SimpleRNN layer.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/SimpleRNNCell,,,,,AIO:Layer -AIO:LSTMCellLayer,Layer,AIO:LayerSubset,,,LSTMCell Layer,0.78,Cell class for the LSTM layer.,A layer that processes one step within the whole time sequence input for an LSTM layer.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/LSTMCell,,,,,AIO:Layer -AIO:DenseFeaturesLayer,Layer,AIO:LayerSubset,,,DenseFeatures Layer,0.86,"A layer that produces a dense Tensor based on given feature_columns. Generally a single example in training data is described with FeatureColumns. At the first layer of the model, this column oriented data should be converted to a single Tensor. This layer can be called multiple times with different features. This is the V2 version of this layer that uses name_scopes to create variables instead of variable_scopes. But this approach currently lacks support for partitioned variables. In that case, use the V1 version instead.",A layer that produces a dense tensor based on given feature columns.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/DenseFeatures,,,,,AIO:Layer -AIO:PoolingLayer,Layer,AIO:LayerSubset,,,Pooling Layer,0.55,Pooling layers serve the dual purposes of mitigating the sensitivity of convolutional layers to location and of spatially downsampling representations.,A layer that serves to mitigate the sensitivity of convolutional layers to location and spatially downsample representations.,https://d2l.ai/chapter_convolutional-neural-networks/pooling.html,,,,,AIO:Layer -AIO:InputSpecLayer,Layer,AIO:LayerSubset,,,InputSpec Layer,0.94,"Specifies the rank, dtype and shape of every input to a layer. Layers can expose (if appropriate) an input_spec attribute: an instance of InputSpec, or a nested structure of InputSpec instances (one per input tensor). These objects enable the layer to run input compatibility checks for input structure, input rank, input shape, and input dtype. A None entry in a shape is compatible with any dimension, a None shape is compatible with any shape.",A layer that specifies the rank,https://www.tensorflow.org/api_docs/python/tf/keras/layers/InputSpec,,,,,AIO:Layer -AIO:EmbeddingLayer,Layer,AIO:LayerSubset,,,Embedding Layer,0.36,Turns positive integers (indexes) into dense vectors of fixed size.,A layer that turns positive integers (indexes) into dense vectors of fixed size.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Embedding,,,,,AIO:Layer -AIO:UpSampling3DLayer,Layer,AIO:LayerSubset,,,UpSampling3D Layer,0.93,Upsampling layer for 3D inputs.,A layer that upsamples the input by repeating each depth,https://www.tensorflow.org/api_docs/python/tf/keras/layers/UpSampling3D,,,,,AIO:Layer -AIO:UpSampling2DLayer,Layer,AIO:LayerSubset,,,UpSampling2D Layer,0.85,Upsampling layer for 2D inputs. Repeats the rows and columns of the data by size[0] and size[1] respectively.,A layer that upsamples the input by repeating each row and column size times.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/UpSampling2D,,,,,AIO:Layer -AIO:LocallyconnectedLayer,Layer,AIO:LayerSubset,,,Locally-connected Layer,,,"A layer that works similarly to the Convolution1D layer, except that weights are unshared, that is, a different set of filters is applied at each different patch of the input.",https://faroit.com/keras-docs/1.2.2/layers/local/,,,,,AIO:Layer -AIO:LambdaLayer,Layer,AIO:LayerSubset,,,Lambda Layer,0.45,Wraps arbitrary expressions as a Layer object.,A layer that wraps arbitrary expressions as a Layer object.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Lambda,,,,,AIO:Layer -AIO:PolicyLayer,Layer,AIO:LayerSubset,,,Policy Layer,,,"A layer that, after taking a set of states or values as input, predicts a probability distribution of actions to take.",,,,,,AIO:Layer -AIO:InputLayerLayer,Layer,AIO:LayerSubset,,,InputLayer Layer,0.18,Layer to be used as an entry point into a Network (a graph of layers).,A layer to be used as an entry point into a Network (a graph of layers).,https://www.tensorflow.org/api_docs/python/tf/keras/layers/InputLayer,,,,,AIO:Layer -AIO:MergingLayer,Layer,AIO:LayerSubset,,,Merging Layer,0.00,A layer used to merge a list of inputs.,A layer used to merge a list of inputs.,https://www.tutorialspoint.com/keras/keras_merge_layer.htm,,,,,AIO:Layer -AIO:WrapperLayer,Layer,AIO:LayerSubset,,,Wrapper Layer,0.83,"Abstract wrapper base class. Wrappers take another layer and augment it in various ways. Do not use this class as a layer, it is only an abstract base class. Two usable wrappers are the TimeDistributed and Bidirectional wrappers.",An abstract base class for wrappers that augment the functionality of another layer.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Wrapper,,,,,AIO:Layer -AIO:AbstractRNNCell,Layer,AIO:LayerSubset,,,AbstractRNNCell,0.36,Abstract object representing an RNN cell. This is the base class for implementing RNN cells with custom behavior.,An abstract layer object representing an RNN cell that is the base class for implementing RNN cells with custom behavior.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/AbstractRNNCell,,,,,AIO:Layer -AIO:RandomBrightnessLayer,Layer,AIO:LayerSubset,,,RandomBrightness Layer,0.81,"A preprocessing layer which randomly adjusts brightness during training. This layer will randomly increase/reduce the brightness for the input RGB images. At inference time, the output will be identical to the input. Call the layer with training=True to adjust the brightness of the input. Note that different brightness adjustment factors will be apply to each the images in the batch.",An image preprocessing layer that randomly adjusts brightness during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomBrightness,,,,,AIO:Layer -AIO:RandomContrastLayer,Layer,AIO:LayerSubset,,,RandomContrast Layer,0.90,"A preprocessing layer which randomly adjusts contrast during training. This layer will randomly adjust the contrast of an image or images by a random factor. Contrast is adjusted independently for each channel of each image during training. For each channel, this layer computes the mean of the image pixels in the channel and then adjusts each component x of each pixel to (x - mean) * contrast_factor + mean. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and in integer or floating point dtype. By default, the layer will output floats. The output value will be clipped to the range [0, 255], the valid range of RGB colors.",An image preprocessing layer that randomly adjusts contrast during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomContrast,,,,,AIO:Layer -AIO:RandomCropLayer,Layer,AIO:LayerSubset,,,RandomCrop Layer,0.90,"A preprocessing layer which randomly crops images during training. During training, this layer will randomly choose a location to crop images down to a target size. The layer will crop all the images in the same batch to the same cropping location. At inference time, and during training if an input image is smaller than the target size, the input will be resized and cropped so as to return the largest possible window in the image that matches the target aspect ratio. If you need to apply random cropping at inference time, set training to True when calling the layer. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats.",An image preprocessing layer that randomly crops images during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomCrop,,,,,AIO:Layer -AIO:RandomFlipLayer,Layer,AIO:LayerSubset,,,RandomFlip Layer,0.87,"A preprocessing layer which randomly flips images during training. This layer will flip the images horizontally and or vertically based on the mode attribute. During inference time, the output will be identical to input. Call the layer with training=True to flip the input. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats.",An image preprocessing layer that randomly flips images during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomFlip,,,,,AIO:Layer -AIO:RandomRotationLayer,Layer,AIO:LayerSubset,,,RandomRotation Layer,0.42,A preprocessing layer which randomly rotates images during training.,An image preprocessing layer that randomly rotates images during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomRotation,,,,,AIO:Layer -AIO:RandomTranslationLayer,Layer,AIO:LayerSubset,,,RandomTranslation Layer,0.84,"A preprocessing layer which randomly translates images during training. This layer will apply random translations to each image during training, filling empty space according to fill_mode. aInput pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats.",An image preprocessing layer that randomly translates images during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomTranslation,,,,,AIO:Layer -AIO:RandomHeightLayer,Layer,AIO:LayerSubset,,,RandomHeight Layer,0.87,"A preprocessing layer which randomly varies image height during training. This layer adjusts the height of a batch of images by a random factor. The input should be a 3D (unbatched) or 4D (batched) tensor in the ""channels_last"" image data format. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. By default, this layer is inactive during inference.",An image preprocessing layer that randomly varies image height during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomHeight,,,,,AIO:Layer -AIO:RandomWidthLayer,Layer,AIO:LayerSubset,,,RandomWidth Layer,0.87,"A preprocessing layer which randomly varies image width during training. This layer will randomly adjusts the width of a batch of images of a batch of images by a random factor. The input should be a 3D (unbatched) or 4D (batched) tensor in the ""channels_last"" image data format. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. By default, this layer is inactive during inference.",An image preprocessing layer that randomly varies image width during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomWidth,,,,,AIO:Layer -AIO:RandomZoomLayer,Layer,AIO:LayerSubset,,,RandomZoom Layer,0.76,"A preprocessing layer which randomly zooms images during training. This layer will randomly zoom in or out on each axis of an image independently, filling empty space according to fill_mode.Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats.",An image preprocessing layer that randomly zooms in or out on images during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomZoom,,,,,AIO:Layer -AIO:LayerLayer,Layer,AIO:LayerSubset,,,Layer Layer,0.93,"This is the class from which all layers inherit. A layer is a callable object that takes as input one or more tensors and that outputs one or more tensors. It involves computation, defined in the call() method, and a state (weight variables). State can be created in various places, at the convenience of the subclass implementer: in __init__(); in the optional build() method, which is invoked by the first __call__() to the layer, and supplies the shape(s) of the input(s), which may not have been known at initialization time; in the first invocation of call(), with some caveats discussed below. Users will just instantiate a layer and then treat it as a callable.",The base class from which all layers inherit.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Layer,,,,,AIO:Layer -AIO:LocallyConnected1DLayer,Layer,AIO:LayerSubset,,,LocallyConnected1D Layer,0.71,"Locally-connected layer for 1D inputs. The LocallyConnected1D layer works similarly to the Conv1D layer, except that weights are unshared, that is, a different set of filters is applied at each different patch of the input.",A locally-connected layer for 1D inputs where each patch of the input is convolved with a different set of filters.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/LocallyConnected1D,,,,,AIO:LocallyconnectedLayer -AIO:LocallyConnected2DLayer,Layer,AIO:LayerSubset,,,LocallyConnected2D Layer,0.71,"Locally-connected layer for 2D inputs. The LocallyConnected2D layer works similarly to the Conv2D layer, except that weights are unshared, that is, a different set of filters is applied at each different patch of the input.",A locally-connected layer for 2D inputs where each patch of the input is convolved with a different set of filters.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/LocallyConnected2D,,,,,AIO:LocallyconnectedLayer -AIO:GatedRecurrentUnit,Network,AIO:NetworkSubset,GRU,,Gated Recurrent Unit,0.43,"Gated recurrent units (GRUs) are a gating mechanism in recurrent neural networks, similar to LSTMs but with fewer parameters and no output gate.",A long short-term memory network that is a gating mechanism in recurrent neural networks similar to LSTMs but with fewer parameters and no output gate.,https://en.wikipedia.org/wiki/Gated_recurrent_unit,"Input, Memory Cell, Output","Layers: Input, Memory Cell, Output",,,AIO:LongShortTermMemory -AIO:Biclustering,Machine Learning,AIO:MachineLearningSubset,Block Clustering|Co-clustering|Two-mode Clustering|Two-way Clustering|Joint Clustering,,Biclustering,0.33,Methods that simultaneously cluster the rows and columns of a matrix to identify submatrices with coherent patterns.,A machine learning task focused on methods that simultaneously cluster the rows and columns of a matrix to identify submatrices with coherent patterns.,https://en.wikipedia.org/wiki/Biclustering|,,,,,AIO:MachineLearning -AIO:Clustering,Machine Learning,AIO:MachineLearningSubset,Cluster analysis,,Clustering,0.30,Methods that group a set of objects such that objects in the same group are more similar to each other than to those in other groups.,A machine learning task focused on methods that group a set of objects such that objects in the same group are more similar to each other than to those in other groups.,https://en.wikipedia.org/wiki/Cluster_analysis|,,,,,AIO:MachineLearning -AIO:KnearestNeighborAlgorithm,Machine Learning,AIO:MachineLearningSubset,KNN|K-NN,,K-nearest Neighbor Algorithm,,,"A machine learning that groups objects by a plurality vote of its neighbors, assigning each object to the class most common among its k nearest neighbors.",https://en.wikipedia.org/wiki/K-nearest_neighbors_algorithm|,,,,,AIO:MachineLearning -AIO:node2vec,Network,AIO:NetworkSubset,N2V|node2vec,,node2vec,,,A machine learning designed to learn continuous feature representations for nodes in a graph by optimizing a neighborhood-preserving objective.,https://en.wikipedia.org/wiki/Node2vec,"Input, Hidden, Output","Layers: Input, Hidden, Output",,AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:MachineLearning -AIO:NaturalLanguageProcessing,Machine Learning,AIO:MachineLearningSubset,NLP,,Natural Language Processing,0.28,"A subfield of linguistics, computer science, and artificial intelligence focused on the interactions between computers and human language, including programming computers to process and analyze large amounts of natural language data.",A subfield of machine learning focused on the interactions between computers and human language including programming computers to process and analyze large amounts of natural language data.,https://en.wikipedia.org/wiki/Natural_language_processing|,,,,,AIO:MachineLearning -AIO:ProbabilisticGraphicalModel,Machine Learning,AIO:MachineLearningSubset,PGM|Graphical Model|Structure Probabilistic Model,,Probabilistic Graphical Model,0.18,A probabilistic model in which a graph expresses the conditional dependence structure between random variables.,A machine learning model in which a graph expresses the conditional dependence structure between random variables.,https://en.wikipedia.org/wiki/Graphical_model|,,,,,AIO:MachineLearning -AIO:ActiveLearning,Machine Learning,AIO:MachineLearningSubset,Query Learning,,Active Learning,0.33,Methods that interactively query a user or another information source to label new data points with the desired outputs.,A type of machine learning focused on methods that interactively query a user or another information source to label new data points with the desired outputs.,https://en.wikipedia.org/wiki/Active_learning_(machine_learning)|,,,,,AIO:MachineLearning -AIO:word2vec,Network,AIO:NetworkSubset,W2V|word2vec,,word2vec,,,"A machine learning that generates distributed representations of words by training a shallow neural network model, which aims to predict the context of each word within a corpus. This algorithm captures semantic meanings of words through their contextual usage in the text.",https://en.wikipedia.org/wiki/Word2vec,"Input, Hidden, Output","Layers: Input, Hidden, Output",,AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:MachineLearning -AIO:SurvivalAnalysis,Machine Learning,AIO:MachineLearningSubset,,,Survival Analysis,0.32,"Methods for analyzing the expected duration of time until one or more events occur, such as death in biological organisms or failure in mechanical systems.",A machine learning task focused on methods for analyzing the expected duration of time until one or more events occur such as death in biological organisms or failure in mechanical systems.,https://en.wikipedia.org/wiki/Survival_analysis|,,,,,AIO:MachineLearning -AIO:TimeSeriesAnalysis,Machine Learning,AIO:MachineLearningSubset,,,Time Series Analysis,0.42,Methods for analyzing time series data to extract meaningful statistics and characteristics.,A machine learning task focused on methods for analyzing time series data to extract meaningful statistics and characteristics.,https://en.wikipedia.org/wiki/Time_series|,,,,,AIO:MachineLearning -AIO:TimeSeriesForecasting,Machine Learning,AIO:MachineLearningSubset,,,Time Series Forecasting,0.44,Methods that predict future values based on previously observed values.,A machine learning task focused on methods that predict future values based on previously observed values.,https://en.wikipedia.org/wiki/Time_series|,,,,,AIO:MachineLearning -AIO:DataImputation,Machine Learning,AIO:MachineLearningSubset,,,Data Imputation,0.53,Methods that replace missing data with substituted values.,A machine learning task focused on methods that replace missing data with substituted values.,https://en.wikipedia.org/wiki/Imputation_(statistics)|,,,,,AIO:MachineLearning -AIO:SelfsupervisedLearning,Machine Learning,AIO:MachineLearningSubset,,,Self-supervised Learning,,,A machine learning that is intermediate between supervised and unsupervised learning and predicts parts of the input data from other observed parts without relying on human-annotated labels.,https://en.wikipedia.org/wiki/Self-supervised_learning|,,,,,AIO:MachineLearning -AIO:UnsupervisedLearning,Machine Learning,AIO:MachineLearningSubset,,,Unsupervised Learning,0.60,Algorithms that learn patterns from unlabeled data.,A type of machine learning focused on algorithms that learn patterns from unlabeled data.,https://en.wikipedia.org/wiki/Unsupervised_learning|,,,,,AIO:MachineLearning -AIO:ReinforcementLearning,Machine Learning,AIO:MachineLearningSubset,,,Reinforcement Learning,0.27,"Methods that do not require labeled input/output pairs or explicit correction of sub-optimal actions, focusing instead on balancing exploration and exploitation to optimize performance over time.",A type of machine learning focused on methods that do not require labeled input/output pairs or explicit correction of sub-optimal actions focusing instead on balancing exploration and exploitation to optimize performance over time.,https://en.wikipedia.org/wiki/Reinforcement_learning|,,,,,AIO:MachineLearning -AIO:SupervisedLearning,Machine Learning,AIO:MachineLearningSubset,,,Supervised Learning,0.38,Methods that learn a function mapping input to output based on example input-output pairs.,A type of machine learning focused on methods that learn a function mapping input to output based on example input-output pairs.,https://en.wikipedia.org/wiki/Supervised_learning|,,,,,AIO:MachineLearning -AIO:EnsembleLearning,Machine Learning,AIO:MachineLearningSubset,,,Ensemble Learning,0.30,Methods that use multiple learning algorithms to achieve better predictive performance than any of the constituent algorithms alone.,A type of machine learning focused on methods that use multiple learning algorithms to achieve better predictive performance than any of the constituent algorithms alone.,https://en.wikipedia.org/wiki/Ensemble_learning|,,,,,AIO:MachineLearning -AIO:MetaLearning,Network,AIO:NetworkSubset,,,Meta-Learning,,,A machine learning that automatically learns from metadata about machine learning experiments.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:MachineLearning -AIO:TransferLearning,Network,AIO:NetworkSubset,,,Transfer Learning,0.30,Methods that reuse or transfer information from previously learned tasks to facilitate the learning of new tasks.,A type of machine learning focused on methods that reuse or transfer information from previously learned tasks to facilitate the learning of new tasks.,https://en.wikipedia.org/wiki/Transfer_learning|,,,,,AIO:MachineLearning -AIO:MultimodalLearning,Network,AIO:NetworkSubset,,,Multimodal Learning,0.42,"A type of deep learning that uses multiple modalities of data, such as text, audio, and images, to improve learning outcomes.",A type of machine learning that uses multiple modalities of data such as text audio and images to improve learning outcomes.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:MachineLearning -AIO:AddLayer,Layer,AIO:LayerSubset,,,Add Layer,0.66,"Layer that adds a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).",A merging layer that adds a list of inputs taking as input a list of tensors all of the same shape.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Add,,,,,AIO:MergingLayer -AIO:AverageLayer,Layer,AIO:LayerSubset,,,Average Layer,0.63,"Layer that averages a list of inputs element-wise. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).",A merging layer that averages a list of inputs element-wise taking as input a list of tensors all of the same shape.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Average,,,,,AIO:MergingLayer -AIO:MaximumLayer,Layer,AIO:LayerSubset,,,Maximum Layer,0.67,"Layer that computes the maximum (element-wise) a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).",A merging layer that computes the maximum (element-wise) of a list of inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Maximum,,,,,AIO:MergingLayer -AIO:MinimumLayer,Layer,AIO:LayerSubset,,,Minimum Layer,0.67,"Layer that computes the minimum (element-wise) a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).",A merging layer that computes the minimum (element-wise) of a list of inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Minimum,,,,,AIO:MergingLayer -AIO:ConcatenateLayer,Layer,AIO:LayerSubset,,,Concatenate Layer,0.56,"Layer that concatenates a list of inputs. It takes as input a list of tensors, all of the same shape except for the concatenation axis, and returns a single tensor that is the concatenation of all inputs.",A merging layer that concatenates a list of inputs taking as input a list of tensors all of the same shape except for the concatenation axis.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Concatenate,,,,,AIO:MergingLayer -AIO:MultiplyLayer,Layer,AIO:LayerSubset,,,Multiply Layer,0.73,"Layer that multiplies (element-wise) a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).",A merging layer that multiplies (element-wise) a list of inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Multiply,,,,,AIO:MergingLayer -AIO:SubtractLayer,Layer,AIO:LayerSubset,,,Subtract Layer,0.87,"Layer that subtracts two inputs. It takes as input a list of tensors of size 2, both of the same shape, and returns a single tensor, (inputs[0] - inputs[1]), also of the same shape.",A merging layer that subtracts two inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Subtract,,,,,AIO:MergingLayer -AIO:AutoregressiveConditionalHeteroskedasticity,Model,AIO:ModelSubset,ARCH,,Autoregressive Conditional Heteroskedasticity,#N/A,,"A model that describes the variance of the current error term as a function of the previous periods' error terms, capturing volatility clustering. Used for time series data.",,,,,,AIO:Model -AIO:AutoregressiveDistributedLag,Model,AIO:ModelSubset,ARDL,,Autoregressive Distributed Lag,#N/A,,"A model that includes lagged values of both the dependent variable and one or more independent variables, capturing dynamic relationships over time. Used in time series analysis.",,,,,,AIO:Model -AIO:AutoregressiveIntegratedMovingAverage,Model,AIO:ModelSubset,ARIMA,,Autoregressive Integrated Moving Average,#N/A,,"A model which combines autoregression (AR), differencing (I), and moving average (MA) components. Used for analyzing and forecasting time series data.",,,,,,AIO:Model -AIO:AutoregressiveMovingAverage,Model,AIO:ModelSubset,ARMA,,Autoregressive Moving Average,#N/A,,"A model that combines autoregressive (AR) and moving average (MA) components to represent time series data, suitable for stationary series without the need for differencing.",,,,,,AIO:Model -AIO:DynamicConditionalCorrelation,Model,AIO:ModelSubset,DCC,,Dynamic Conditional Correlation,#N/A,,"A model that allows for time-varying correlations between different time series, used in financial econometrics to model and forecast covariances.",,,,,,AIO:Model -AIO:ExponentialSmoothingStateSpaceModel,Model,AIO:ModelSubset,ETS,,Exponential Smoothing State Space Model,#N/A,,"A model that combines exponential smoothing with state space modeling, allowing for the inclusion of both trend and seasonal components. Used in forecasting.",,,,,,AIO:Model -AIO:GeneralizedAutoregressiveConditionalHeteroskedasticity,Model,AIO:ModelSubset,GARCH,,Generalized Autoregressive Conditional Heteroskedasticity,#N/A,,"A model that incorporates lagged conditional variances, allowing for more flexibility in modeling time-varying volatility.",,,,,,AIO:Model -AIO:LanguageModel,Model,AIO:ModelSubset,Language Model,,Language Model,0.09,A language model is a probabilistic model designed to predict the next word in a sequence or assign probabilities to sequences of words in natural language.,A model designed to predict the next word in a sequence or assign probabilities to sequences of words in natural language.,https://en.wikipedia.org/wiki/Language_model|,,,,,AIO:Model -AIO:SeasonalAutoregressiveIntegratedMovingAverage,Model,AIO:ModelSubset,SARIMA,,Seasonal Autoregressive Integrated Moving-Average,,,"A model that extends ARIMA, explicitly supporting univariate time series data with a seasonal component, combining seasonal differencing with ARIMA modeling.",,,,,,AIO:Model -AIO:ThresholdAutoregressive,Model,AIO:ModelSubset,TAR,,Threshold Autoregressive,#N/A,,"A model that allows for different autoregressive processes depending on the regime or state of the time series, enabling the capture of nonlinear behaviors.",,,,,,AIO:Model -AIO:VectorAutoregression,Model,AIO:ModelSubset,VAR,,Vector Autoregression,#N/A,,"A model that captures the linear interdependencies among multiple time series, where each variable is modeled as a linear function of its own past values and the past values of all other variables in the system.",,,,,,AIO:Model -AIO:ModularLLM,Model,AIO:ModelSubset,Modular Large Language Model,component skills|skill composition,Modular LLM,0.20,,A modular large language model that consists of multiple specialized components or skills that can be dynamically composed and recombined to solve complex tasks mimicking the modular structure of human cognition.,https://arxiv.org/abs/2302.11529v2|,,,,,AIO:ModularLanguageModel -AIO:MultimodalLLM,Model,AIO:ModelSubset,Multimodal Large Language Model,cross-modal grounding,Multimodal LLM,0.34,,A multimodal large language model that learns joint representations across different modalities like text vision and audio in an end-to-end fashion for better cross-modal understanding and generation.,https://arxiv.org/abs/2303.17580|,,,,,AIO:MultimodalLanguageModel -AIO:MultimodalPromptbasedLanguageModel,Model,AIO:ModelSubset,Multimodal Prompt-based Language Model,,Multimodal Prompt-based Language Model,,,"A multimodal LLM which processes prompts that include multiple modalities, such as both text and images, to generate relevant responses.",https://arxiv.org/abs/2210.03094|,,,,,AIO:MultimodalLanguageModel -AIO:ArtificialNeuralNetwork,Network,AIO:NetworkSubset,ANN|NN,,Artificial Neural Network,0.58,"An artificial neural network (ANN) is based on a collection of connected units or nodes called artificial neurons, modeled after biological neurons, with connections transmitting signals processed by non-linear functions.",A network based on a collection of connected units called artificial neurons modeled after biological neurons.,https://en.wikipedia.org/wiki/Artificial_neural_network,,,,,AIO:Network -AIO:KohonenNetwork,Network,AIO:NetworkSubset,KN|Self-Organizing Map|SOM|Self-Organizing Feature Map|SOFM,,Kohonen Network,0.40,"A self-organizing map (SOM) or Kohonen network is an unsupervised machine learning technique producing a low-dimensional representation of high-dimensional data, preserving topological structure.",A network that is an unsupervised technique producing a low-dimensional representation of high-dimensional data preserving topological structure.,https://en.wikipedia.org/wiki/Self-organizing_map,"Input, Hidden","Layers: Input, Hidden",,AIO:InputLayer|AIO:HiddenLayer,AIO:Network -AIO:LiquidStateMachineNetwork,Network,AIO:NetworkSubset,LSM,,Liquid State Machine Network,0.50,"A liquid state machine (LSM) is a type of reservoir computer using a spiking neural network, with recurrently connected nodes turning time-varying input into spatio-temporal activation patterns.",A network that is a type of reservoir computer turning time-varying input into spatio-temporal activation patterns.,https://en.wikipedia.org/wiki/Liquid_state_machine,"Input, Spiking Hidden, Output","Layers: Input, Spiking Hidden, Output",,,AIO:Network -AIO:MarkovChain,Model,AIO:ModelSubset,MC|Markov Process|MP,,Markov Chain,0.16,A Markov chain is a stochastic model describing a sequence of possible events where the probability of each event depends only on the previous event's state.,A network that is a stochastic model describing a sequence of possible events where the probability of each event depends only on the previous event's state.,https://en.wikipedia.org/wiki/Markov_chain,Probalistic Hidden,Layers: Probalistic Hidden,,,AIO:Network -AIO:SymmetricallyConnectedNetwork,Network,AIO:NetworkSubset,SCN,,Symmetrically Connected Network,0.59,"Symmetrically connected networks are a type of recurrent neural network where connections between units are symmetrical, meaning they have equal weights in both directions. This structure allows the network to maintain consistent information flow and equilibrium.",A network that is a type of recurrent neural network where connections between units are symmetrical with equal weights in both directions.,https://ieeexplore.ieee.org/document/287176,,,,,AIO:Network -AIO:SupportVectorMachine,Network,AIO:NetworkSubset,SVM|Supper Vector Network|SVN,,Support Vector Machine,0.45,"Support vector machines (SVMs) are supervised learning models for classification and regression analysis, mapping training examples to points in space to maximize the gap between categories.",A network with supervised learning models for classification and regression that maps training examples to points in space maximizing the gap between categories.,https://en.wikipedia.org/wiki/Support-vector_machine,"Input, Hidden, Output","Layers: Input, Hidden, Output",,AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:Network -AIO:UnsupervisedPretrainedNetwork,Network,AIO:NetworkSubset,UPN,,Unsupervised Pretrained Network,0.54,"Unsupervised pre-training initializes a discriminative neural net from one trained using an unsupervised criterion, aiding in optimization and overfitting issues.",A network that initializes a discriminative neural net from one trained using an unsupervised criterion.,"https://metacademy.org/graphs/concepts/unsupervised_pre_training#:~:text=Unsupervised%20pre%2Dtraining%20initializes%20a,optimization%20and%20the%20overfitting%20issues",,,,,AIO:Network -AIO:BayesianNetwork,Network,AIO:NetworkSubset,,,Bayesian Network,0.30,A probabilistic graphical model representing variables and their conditional dependencies via a directed acyclic graph (DAG).,A network that is a probabilistic graphical model representing variables and their conditional dependencies via a directed acyclic graph.,https://en.wikipedia.org/wiki/Bayesian_network,,,,,AIO:Network -AIO:RecurrentNeuralNetwork,Network,AIO:NetworkSubset,RecNN|Recurrent Network|RN,,Recurrent Neural Network,0.35,"A recurrent neural network (RNN) has connections forming a directed graph along a temporal sequence, enabling dynamic temporal behavior.",A deep neural network with connections forming a directed graph along a temporal sequence enabling dynamic behavior.,,,,,,AIO:Network -AIO:node2vecSkipGram,Network,AIO:NetworkSubset,N2V-SkipGram,SkipGram,node2vec-SkipGram,,,"A node2vec that uses the current node to predict the surrounding window of context nodes, weighing nearby context nodes more heavily than distant ones.",https://en.wikipedia.org/wiki/Word2vec,"Input, Hidden, Output","Layers: Input, Hidden, Output",,AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:node2vec -AIO:GroupNormLayer,Layer,AIO:LayerSubset,GroupNorm|GroupNorm,,GroupNorm Layer,0.68,Applies Group Normalization over a mini-batch of inputs as described in the paper Group Normalization,A normalization layer that applies Group Normalization over a mini-batch of inputs.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:NormalizationLayer -AIO:InstanceNorm1dLayer,Layer,AIO:LayerSubset,InstanceNorm1d|InstanceNorm1d|InstanceNorm1D,,InstanceNorm1d Layer,0.69,Applies Instance Normalization over a 2D (unbatched) or 3D (batched) input as described in the paper Instance Normalization: The Missing Ingredient for Fast Stylization.,A normalization layer that applies Instance Normalization over a 2D (unbatched) or 3D (batched) input.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:NormalizationLayer -AIO:InstanceNorm2d,Layer,AIO:LayerSubset,InstanceNorm2d|InstanceNorm2d|InstanceNorm2D,,InstanceNorm2d,0.59,Applies Instance Normalization over a 4D input (a mini-batch of 2D inputs with additional channel dimension) as described in the paper Instance Normalization: The Missing Ingredient for Fast Stylization.,A normalization layer that applies Instance Normalization over a 4D input (a mini-batch of 2D inputs with additional channel dimension).,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:NormalizationLayer -AIO:InstanceNorm3dLayer,Layer,AIO:LayerSubset,InstanceNorm3d|InstanceNorm3d|InstanceNorm3D,,InstanceNorm3d Layer,0.59,Applies Instance Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension) as described in the paper Instance Normalization: The Missing Ingredient for Fast Stylization.,A normalization layer that applies Instance Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension).,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:NormalizationLayer -AIO:LayerNormLayer,Layer,AIO:LayerSubset,LayerNorm|LayerNorm,,LayerNorm Layer,0.68,Applies Layer Normalization over a mini-batch of inputs as described in the paper Layer Normalization,A normalization layer that applies Layer Normalization over a mini-batch of inputs.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:NormalizationLayer -AIO:LazyInstanceNorm1dLayer,Layer,AIO:LayerSubset,LazyInstanceNorm1d|LazyInstanceNorm1d|LazyInstanceNorm1D,,LazyInstanceNorm1d Layer,0.81,A torch.nn.InstanceNorm1d module with lazy initialization of the num_features argument of the InstanceNorm1d that is inferred from the input.size(1).,An instance normalization layer that lazily initializes the num_features argument from the input size for 1D data.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:NormalizationLayer -AIO:LazyInstanceNorm2dLayer,Layer,AIO:LayerSubset,LazyInstanceNorm2d|LazyInstanceNorm2d|LazyInstanceNorm2D,,LazyInstanceNorm2d Layer,0.81,A torch.nn.InstanceNorm2d module with lazy initialization of the num_features argument of the InstanceNorm2d that is inferred from the input.size(1).,An instance normalization layer that lazily initializes the num_features argument from the input size for 2D data.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:NormalizationLayer -AIO:LazyInstanceNorm3dLayer,Layer,AIO:LayerSubset,LazyInstanceNorm3d|LazyInstanceNorm3d|LazyInstanceNorm3D,,LazyInstanceNorm3d Layer,0.81,A torch.nn.InstanceNorm3d module with lazy initialization of the num_features argument of the InstanceNorm3d that is inferred from the input.size(1).,An instance normalization layer that lazily initializes the num_features argument from the input size for 3D data.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:NormalizationLayer -AIO:LocalResponseNormLayer,Layer,AIO:LayerSubset,LocalResponseNorm|LocalResponseNorm,,LocalResponseNorm Layer,0.57,"Applies local response normalization over an input signal composed of several input planes, where channels occupy the second dimension.",A normalization layer that applies local response normalization over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,,,AIO:NormalizationLayer -AIO:LayerNormalizationLayer,Layer,AIO:LayerSubset,,,LayerNormalization Layer,0.89,"Layer normalization layer (Ba et al., 2016). Normalize the activations of the previous layer for each given example in a batch independently, rather than across a batch like Batch Normalization. i.e. applies a transformation that maintains the mean activation within each example close to 0 and the activation standard deviation close to 1. Given a tensor inputs, moments are calculated and normalization is performed across the axes specified in axis.",A normalization layer that applies Layer Normalization over the inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/LayerNormalization,,,,,AIO:NormalizationLayer -AIO:BatchNormalizationLayer,Layer,AIO:LayerSubset,,,BatchNormalization Layer,0.87,"Layer that normalizes its inputs. Batch normalization applies a transformation that maintains the mean output close to 0 and the output standard deviation close to 1. Importantly, batch normalization works differently during training and during inference. During training (i.e. when using fit() or when calling the layer/model with the argument training=True), the layer normalizes its output using the mean and standard deviation of the current batch of inputs. That is to say, for each channel being normalized, the layer returns gamma * (batch - mean(batch)) / sqrt(var(batch) + epsilon) + beta, where: epsilon is small constant (configurable as part of the constructor arguments), gamma is a learned scaling factor (initialized as 1), which can be disabled by passing scale=False to the constructor. beta is a learned offset factor (initialized as 0), which can be disabled by passing center=False to the constructor. During inference (i.e. when using evaluate() or predict() or when calling the layer/model with the argument training=False (which is the default), the layer normalizes its output using a moving average of the mean and standard deviation of the batches it has seen during training. That is to say, it returns gamma * (batch - self.moving_mean) / sqrt(self.moving_var + epsilon) + beta. self.moving_mean and self.moving_var are non-trainable variables that are updated each time the layer in called in training mode, as such: moving_mean = moving_mean * momentum + mean(batch) * (1 - momentum) moving_var = moving_var * momentum + var(batch) * (1 - momentum).",A normalization layer that normalizes its inputs applying a transformation that maintains the mean close to 0 and the standard deviation close to 1.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/BatchNormalization,,,,,AIO:NormalizationLayer -AIO:NormalizationLayer,Layer,AIO:LayerSubset,,,Normalization Layer,0.25,A preprocessing layer which normalizes continuous features.,A preprocessing layer that normalizes continuous features.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Normalization,,,,,AIO:NumericalFeaturesPreprocessingLayer -AIO:DiscretizationLayer,Layer,AIO:LayerSubset,,,Discretization Layer,0.00,A preprocessing layer which buckets continuous features by ranges.,A preprocessing layer which buckets continuous features by ranges.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Discretization,,,,,AIO:NumericalFeaturesPreprocessingLayer -AIO:AdaptiveAvgPool1DLayer,Layer,AIO:LayerSubset,AdaptiveAvgPool1d|AdaptiveAvgPool1D,,AdaptiveAvgPool1D Layer,0.28,Applies a 1D adaptive average pooling over an input signal composed of several input planes.,A pooling layer that applies a 1D adaptive average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:AdaptiveAvgPool2DLayer,Layer,AIO:LayerSubset,AdaptiveAvgPool2d|AdaptiveAvgPool2D,,AdaptiveAvgPool2D Layer,0.28,Applies a 2D adaptive average pooling over an input signal composed of several input planes.,A pooling layer that applies a 2D adaptive average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:AdaptiveAvgPool3DLayer,Layer,AIO:LayerSubset,AdaptiveAvgPool3d|AdaptiveAvgPool3D,,AdaptiveAvgPool3D Layer,0.28,Applies a 3D adaptive average pooling over an input signal composed of several input planes.,A pooling layer that applies a 3D adaptive average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:AdaptiveMaxPool1DLayer,Layer,AIO:LayerSubset,AdaptiveMaxPool1d|AdaptiveMaxPool1D,,AdaptiveMaxPool1D Layer,0.28,Applies a 1D adaptive max pooling over an input signal composed of several input planes.,A pooling layer that applies a 1D adaptive max pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:AdaptiveMaxPool2DLayer,Layer,AIO:LayerSubset,AdaptiveMaxPool2d|AdaptiveMaxPool2D,,AdaptiveMaxPool2D Layer,0.28,Applies a 2D adaptive max pooling over an input signal composed of several input planes.,A pooling layer that applies a 2D adaptive max pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:AdaptiveMaxPool3DLayer,Layer,AIO:LayerSubset,AdaptiveMaxPool3d|AdaptiveMaxPool3D,,AdaptiveMaxPool3D Layer,0.28,Applies a 3D adaptive max pooling over an input signal composed of several input planes.,A pooling layer that applies a 3D adaptive max pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:AvgPool1DLayer,Layer,AIO:LayerSubset,AvgPool1d|AvgPool1D,,AvgPool1D Layer,0.29,Applies a 1D average pooling over an input signal composed of several input planes.,A pooling layer that applies a 1D average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:AveragePooling1DLayer,Layer,AIO:LayerSubset,AvgPool1D|AvgPool1d,,AveragePooling1D Layer,0.90,"Average pooling for temporal data. Downsamples the input representation by taking the average value over the window defined by pool_size. The window is shifted by strides. The resulting output when using ""valid"" padding option has a shape of: output_shape = (input_shape - pool_size + 1) / strides). The resulting output shape when using the ""same"" padding option is: output_shape = input_shape / strides.",A pooling layer that performs average pooling for temporal data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/AveragePooling1D,,,,,AIO:PoolingLayer -AIO:AvgPool2DLayer,Layer,AIO:LayerSubset,AvgPool2d|AvgPool2D,,AvgPool2D Layer,0.29,Applies a 2D average pooling over an input signal composed of several input planes.,A pooling layer that applies a 2D average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:AveragePooling2DLayer,Layer,AIO:LayerSubset,AvgPool2D|AvgPool2d,,AveragePooling2D Layer,0.93,"Average pooling operation for spatial data. Downsamples the input along its spatial dimensions (height and width) by taking the average value over an input window (of size defined by pool_size) for each channel of the input. The window is shifted by strides along each dimension. The resulting output when using ""valid"" padding option has a shape (number of rows or columns) of: output_shape = math.floor((input_shape - pool_size) / strides) + 1 (when input_shape >= pool_size). The resulting output shape when using the ""same"" padding option is: output_shape = math.floor((input_shape - 1) / strides) + 1.",A pooling layer that performs average pooling for spatial data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/AveragePooling2D,,,,,AIO:PoolingLayer -AIO:AvgPool3DLayer,Layer,AIO:LayerSubset,AvgPool3d|AvgPool3D,,AvgPool3D Layer,0.29,Applies a 3D average pooling over an input signal composed of several input planes.,A pooling layer that applies a 3D average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:AveragePooling3DLayer,Layer,AIO:LayerSubset,AvgPool3D|AvgPool3D,,AveragePooling3D Layer,0.82,"Average pooling operation for 3D data (spatial or spatio-temporal). Downsamples the input along its spatial dimensions (depth, height, and width) by taking the average value over an input window (of size defined by pool_size) for each channel of the input. The window is shifted by strides along each dimension.",A pooling layer that performs average pooling for 3D data (spatial or spatio-temporal).,https://www.tensorflow.org/api_docs/python/tf/keras/layers/AveragePooling3D,,,,,AIO:PoolingLayer -AIO:FractionalMaxPool2DLayer,Layer,AIO:LayerSubset,FractionalMaxPool2d|FractionalMaxPool2D,,FractionalMaxPool2D Layer,0.28,Applies a 2D fractional max pooling over an input signal composed of several input planes.,A pooling layer that applies a 2D fractional max pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:FractionalMaxPool3DLayer,Layer,AIO:LayerSubset,FractionalMaxPool3d|FractionalMaxPool3D,,FractionalMaxPool3D Layer,0.28,Applies a 3D fractional max pooling over an input signal composed of several input planes.,A pooling layer that applies a 3D fractional max pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:GlobalAveragePooling1DLayer,Layer,AIO:LayerSubset,GlobalAvgPool1D|GlobalAvgPool1D,,GlobalAveragePooling1D Layer,0.50,Global average pooling operation for temporal data.,A pooling layer that performs global average pooling operation for temporal data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalAveragePooling1D,,,,,AIO:PoolingLayer -AIO:GlobalAveragePooling2DLayer,Layer,AIO:LayerSubset,GlobalAvgPool2D|GlobalAvgPool2D,,GlobalAveragePooling2D Layer,0.50,Global average pooling operation for spatial data.,A pooling layer that performs global average pooling operation for spatial data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalAveragePooling2D,,,,,AIO:PoolingLayer -AIO:GlobalAveragePooling3DLayer,Layer,AIO:LayerSubset,GlobalAvgPool3D|GlobalAvgPool3D,,GlobalAveragePooling3D Layer,0.62,Global Average pooling operation for 3D data.,A pooling layer that performs global average pooling operation for 3D data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalAveragePooling3D,,,,,AIO:PoolingLayer -AIO:GlobalMaxPooling1DLayer,Layer,AIO:LayerSubset,GlobalMaxPool1D|GlobalMaxPool1D,,GlobalMaxPooling1D Layer,0.54,Global max pooling operation for 1D temporal data.,A pooling layer that performs global max pooling operation for temporal data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalMaxPool1D,,,,,AIO:PoolingLayer -AIO:GlobalMaxPooling2DLayer,Layer,AIO:LayerSubset,GlobalMaxPool2D|GlobalMaxPool2D,,GlobalMaxPooling2D Layer,0.50,Global max pooling operation for spatial data.,A pooling layer that performs global max pooling operation for spatial data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalMaxPool2D,,,,,AIO:PoolingLayer -AIO:GlobalMaxPooling3DLayer,Layer,AIO:LayerSubset,GlobalMaxPool3D|GlobalMaxPool3D,,GlobalMaxPooling3D Layer,0.62,Global Max pooling operation for 3D data.,A pooling layer that performs global max pooling operation for 3D data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalMaxPool3D,,,,,AIO:PoolingLayer -AIO:LPPool1DLayer,Layer,AIO:LayerSubset,LPPool1d|LPPool1D,,LPPool1D Layer,0.35,Applies a 1D power-average pooling over an input signal composed of several input planes.,A pooling layer that applies 1D power-average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:LPPool2DLayer,Layer,AIO:LayerSubset,LPPool2d|LPPool2D,,LPPool2D Layer,0.35,Applies a 2D power-average pooling over an input signal composed of several input planes.,A pooling layer that applies 2D power-average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:MaxPooling1DLayer,Layer,AIO:LayerSubset,MaxPool1D|MaxPool1d|MaxPooling1D|MaxPooling1d,,MaxPooling1D Layer,0.91,"Max pooling operation for 1D temporal data. Downsamples the input representation by taking the maximum value over a spatial window of size pool_size. The window is shifted by strides. The resulting output, when using the ""valid"" padding option, has a shape of: output_shape = (input_shape - pool_size + 1) / strides) The resulting output shape when using the ""same"" padding option is: output_shape = input_shape / strides.",A pooling layer that performs max pooling operation for temporal data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/MaxPool1D,,,,,AIO:PoolingLayer -AIO:MaxPooling2DLayer,Layer,AIO:LayerSubset,MaxPool2D|MaxPool2d|MaxPooling2D|MaxPooling2d,,MaxPooling2D Layer,0.58,Max pooling operation for 2D spatial data.,A pooling layer that performs max pooling operation for spatial data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/MaxPool2D,,,,,AIO:PoolingLayer -AIO:MaxPooling3DLayer,Layer,AIO:LayerSubset,MaxPool3D|MaxPool3d|MaxPooling3D|MaxPooling3d,,MaxPooling3D Layer,0.82,"Max pooling operation for 3D data (spatial or spatio-temporal). Downsamples the input along its spatial dimensions (depth, height, and width) by taking the maximum value over an input window (of size defined by pool_size) for each channel of the input. The window is shifted by strides along each dimension.",A pooling layer that performs max pooling operation for 3D data (spatial or spatio-temporal).,https://www.tensorflow.org/api_docs/python/tf/keras/layers/MaxPool3D,,,,,AIO:PoolingLayer -AIO:MaxUnpool1DLayer,Layer,AIO:LayerSubset,MaxUnpool1d|MaxUnpool1D,,MaxUnpool1D Layer,0.55,Computes a partial inverse of MaxPool1d.,A pooling layer that computes a partial inverse of MaxPool1d.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:MaxUnpool2DLayer,Layer,AIO:LayerSubset,MaxUnpool2d|MaxUnpool2D,,MaxUnpool2D Layer,0.55,Computes a partial inverse of MaxPool2d.,A pooling layer that computes a partial inverse of MaxPool2d.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:MaxUnpool3DLayer,Layer,AIO:LayerSubset,MaxUnpool3d|MaxUnpool3D,,MaxUnpool3D Layer,0.55,Computes a partial inverse of MaxPool3d.,A pooling layer that computes a partial inverse of MaxPool3d.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,,,AIO:PoolingLayer -AIO:Distillation,Preprocessing,AIO:PreprocessingSubset,Refining|Purification,Knowledge compression|Teacher-student model,Distillation,0.04,"The process of training a smaller model to replicate the behavior of a larger model, aiming to compress the knowledge into a more compact form without significant loss of performance.",The process of training a smaller model to replicate the behavior of a larger model aiming to compress the knowledge into a more compact form without significant loss of performance.,https://doi.org/10.48550/arXiv.2105.13093|,,,,,AIO:Preprocessing -AIO:DataPreparation,Preprocessing,AIO:PreprocessingSubset,Data Curation|Data Processing|Data Assembly,,Data Preparation,0.23,"The process of cleaning, transforming, and organizing raw data into a suitable format for analysis and modeling, ensuring the quality and relevance of the data for machine learning tasks.",The process of cleaning transforming and organizing raw data into a suitable format for analysis and modeling ensuring the quality and relevance of the data for machine learning tasks.,,,,,,AIO:Preprocessing -AIO:TrainingStrategies,Preprocessing,AIO:PreprocessingSubset,Learning Techniques|Instructional Methods,,Training Strategies,0.12,"The methodologies and approaches used to train machine learning models, including techniques such as supervised learning, unsupervised learning, reinforcement learning, and transfer learning, aimed at optimizing model performance.",The methodologies and approaches used to train machine learning models including techniques such as supervised learning unsupervised learning reinforcement learning and transfer learning aimed at optimizing model performance.,,,,,,AIO:Preprocessing -AIO:ModelEfficiency,Preprocessing,AIO:PreprocessingSubset,Model Optimization|Computational Efficiency,,Model Efficiency,0.17,"Techniques aimed at making models more efficient, such as knowledge distillation.",Techniques aimed at making models more efficient such as knowledge distillation.,https://doi.org/10.1145/3578938|,,,,,AIO:Preprocessing -AIO:DataEnhancement,Preprocessing,AIO:PreprocessingSubset,,,DataEnhancement,0.38,"Techniques used to improve the quality, diversity, and volume of data available for training machine learning models, such as data augmentation, synthesis, and enrichment, to enhance model robustness and accuracy.",Techniques used to improve the quality diversity and volume of data available for training machine learning models such as data augmentation synthesis and enrichment to enhance model robustness and accuracy.,,,,,,AIO:Preprocessing -AIO:CausalGraphicalModel,Machine Learning,AIO:MachineLearningSubset,Casaul Graph|Path Diagram|Casaul Bayesian Network|DAG|Directed Acyclic Graph,,Causal Graphical Model,0.36,Probabilistic graphical models used to encode assumptions about the data-generating process.,A probabilistic graphical model used to encode assumptions about the data-generating process.,https://en.wikipedia.org/wiki/Causal_graph|,,,,,AIO:ProbabilisticGraphicalModel -AIO:ProbabilisticTopicModel,Machine Learning,AIO:MachineLearningSubset,,,Probabilistic Topic Model,0.39,"Methods that use statistical techniques to analyze the words in each text to discover common themes, their connections, and their changes over time.",A probabilistic graphical model that uses statistical techniques to analyze the words in each text to discover common themes their connections and their changes over time.,https://pyro.ai/examples/prodlda.html|,,,,,AIO:ProbabilisticGraphicalModel -AIO:ErrorPropagationBias,Bias,AIO:BiasSubset,Error Propagation,,Error Propagation Bias,0.33,"The effect of variables' uncertainties (or errors, more specifically random errors) on the uncertainty of a function based on them.",A processing bias characterized by the effect of variables' uncertainties (or errors more specifically random errors) on the uncertainty of a function based on them.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:ProcessingBias -AIO:InheritedBias,Bias,AIO:BiasSubset,,,Inherited Bias,0.30,"Bias arising when machine learning applications generate inputs for other machine learning algorithms, passing on any existing bias.",A processing bias arising when machine learning applications generate inputs for other machine learning algorithms passing on any existing bias.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:ProcessingBias -AIO:AmplificationBias,Bias,AIO:BiasSubset,,,Amplification Bias,0.22,Bias arising when the distribution over prediction outputs is skewed compared to the prior distribution of the prediction target.,A processing bias arising when the distribution over prediction outputs is skewed compared to the prior distribution of the prediction target.,https://royalsocietypublishing.org/doi/10.1098/rspb.2019.0165#d1e5237|GTP-4o with Seppala et al. 2017,,,,,AIO:ProcessingBias -AIO:SurvivorshipBias,Bias,AIO:BiasSubset,,,Survivorship Bias,0.48,"The tendency to focus on items, observations, or people that ""survive"" a selection process, overlooking those that did not.","A processing bias characterized by the tendency to focus on items observations or people that ""survive"" a selection process overlooking those that did not.",https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:ProcessingBias -AIO:ModelSelectionBias,Bias,AIO:BiasSubset,,,Model Selection Bias,0.21,"Bias introduced when using data to select a single ""best"" model from many, or when an explanatory variable has a weak relationship with the response variable.","A processing bias introduced when using data to select a single ""best"" model from many or when an explanatory variable has a weak relationship with the response variable.",https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:ProcessingBias -AIO:UnitNormalizationLayer,Layer,AIO:LayerSubset,,,UnitNormalization Layer,0.43,Unit normalization layer. Normalize a batch of inputs so that each input in the batch has a L2 norm equal to 1 (across the axes specified in axis).,A normalization layer that normalizes a batch of inputs so that each input in the batch has a L2 norm equal to 1.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/UnitNormalization,,,,,AIO:RecurrentLayer -AIO:SimpleRNNLayer,Layer,AIO:LayerSubset,,,SimpleRNN Layer,0.44,Fully-connected RNN where the output is to be fed back to input.,A recurrent layer that implements a fully-connected RNN where the output is to be fed back to input.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/SimpleRNN,,,,,AIO:RecurrentLayer -AIO:GRULayer,Layer,AIO:LayerSubset,,,GRU Layer,0.96,"Gated Recurrent Unit - Cho et al. 2014. Based on available runtime hardware and constraints, this layer will choose different implementations (cuDNN-based or pure-TensorFlow) to maximize the performance. If a GPU is available and all the arguments to the layer meet the requirement of the cuDNN kernel (see below for details), the layer will use a fast cuDNN implementation. The requirements to use the cuDNN implementation are: activation == tanh, recurrent_activation == sigmoid, recurrent_dropout == 0, unroll is False, use_bias is True, reset_after is True. Inputs, if use masking, are strictly right-padded. Eager execution is enabled in the outermost context. There are two variants of the GRU implementation. The default one is based on v3 and has reset gate applied to hidden state before matrix multiplication. The other one is based on original and has the order reversed. The second variant is compatible with CuDNNGRU (GPU-only) and allows inference on CPU. Thus it has separate biases for kernel and recurrent_kernel. To use this variant, set reset_after=True and recurrent_activation='sigmoid'.",A recurrent layer that implements the Gated Recurrent Unit architecture.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GRU,,,,,AIO:RecurrentLayer -AIO:LSTMLayer,Layer,AIO:LayerSubset,,,LSTM Layer,0.94,"Long Short-Term Memory layer - Hochreiter 1997. Based on available runtime hardware and constraints, this layer will choose different implementations (cuDNN-based or pure-TensorFlow) to maximize the performance. If a GPU is available and all the arguments to the layer meet the requirement of the cuDNN kernel (see below for details), the layer will use a fast cuDNN implementation. The requirements to use the cuDNN implementation are: 1. activation == tanh, 2. recurrent_activation == sigmoid, 3. recurrent_dropout == 0, 4. unroll is False, 5. use_bias is True, 6. Inputs, if use masking, are strictly right-padded, 7. Eager execution is enabled in the outermost context.",A recurrent layer that implements the Long Short-Term Memory architecture.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/LSTM,,,,,AIO:RecurrentLayer -AIO:BidirectionalLayer,Layer,AIO:LayerSubset,,,Bidirectional Layer,0.73,Bidirectional wrapper for RNNs.,A recurrent layer that is a bidirectional wrapper for RNNs.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Bidirectional,,,,,AIO:RecurrentLayer -AIO:TimeDistributedLayer,Layer,AIO:LayerSubset,,,TimeDistributed Layer,0.85,"This wrapper allows to apply a layer to every temporal slice of an input. Every input should be at least 3D, and the dimension of index one of the first input will be considered to be the temporal dimension. Consider a batch of 32 video samples, where each sample is a 128x128 RGB image with channels_last data format, across 10 timesteps. The batch input shape is (32, 10, 128, 128, 3). You can then use TimeDistributed to apply the same Conv2D layer to each of the 10 timesteps, independently:",A wrapper layer that applies a layer to every temporal slice of an input.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/TimeDistributed,,,,,AIO:RecurrentLayer -AIO:RNNLayer,Layer,AIO:LayerSubset,,,RNN Layer,0.43,Base class for recurrent layers.,The base class for recurrent layers.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RNN,,,,,AIO:RecurrentLayer -AIO:EchoStateNetwork,Network,AIO:NetworkSubset,ESN,,Echo State Network,0.53,"An echo state network (ESN) is a type of reservoir computer with a recurrent neural network and a sparsely connected hidden layer, learning output neuron weights to produce temporal patterns.",A recurrent neural network with a recurrent hidden layer and sparsely connected hidden neurons that learns output weights to produce temporal patterns.,"https://en.wikipedia.org/wiki/Echo_state_network#:~:text=The%20echo%20state%20network%20(ESN,are%20fixed%20and%20randomly%20assigned","Input, Recurrent, Output","Layers: Input, Recurrent, Output",,,AIO:RecurrentNeuralNetwork -AIO:LongShortTermMemory,Network,AIO:NetworkSubset,LSTM,,Long Short Term Memory,0.77,"Long short-term memory (LSTM) networks are artificial recurrent neural networks with feedback connections, processing entire sequences of data for tasks like handwriting and speech recognition.",A recurrent neural network with feedback connections that processes entire sequences of data.,https://en.wikipedia.org/wiki/Long_short-term_memory,"Input, Memory Cell, Output","Layers: Input, Memory Cell, Output",,,AIO:RecurrentNeuralNetwork -AIO:FixedEffectsModel,Machine Learning,AIO:MachineLearningSubset,FEM,,Fixed Effects Model,0.21,A statistical model in which the model parameters are fixed or non-random quantities.,A regression analysis model in which the model parameters are fixed or non-random quantities.,https://en.wikipedia.org/wiki/Fixed_effects_model|,,,,,AIO:RegressionAnalysis -AIO:GeneralizedLinearModel,Machine Learning,AIO:MachineLearningSubset,GLM,,Generalized Linear Model,0.07,A model that generalizes linear regression by relating the linear model to the response variable via a link function and allowing the variance of each measurement to be a function of its predicted value.,A machine learning model that generalizes linear regression by relating the linear model to the response variable via a link function and allowing the variance of each measurement to be a function of its predicted value.,https://en.wikipedia.org/wiki/Generalized_linear_model|,,,,,AIO:RegressionAnalysis -AIO:KnearestNeighborRegressionAlgorithm,Machine Learning,AIO:MachineLearningSubset,KNN|K-NN,,K-nearest Neighbor Regression Algorithm,,,An regression analysis that assigns the average of the values of k nearest neighbors to objects.,https://en.wikipedia.org/wiki/K-nearest_neighbors_algorithm|,,,,,AIO:RegressionAnalysis -AIO:RandomEffectsModel,Machine Learning,AIO:MachineLearningSubset,REM,,Random Effects Model,0.27,A statistical model where the model parameters are random variables.,A regression analysis model where the model parameters are random variables.,https://en.wikipedia.org/wiki/Random_effects_model|,,,,,AIO:RegressionAnalysis -AIO:RidgeRegression,Machine Learning,AIO:MachineLearningSubset,,,Ridge Regression,0.20,A method of estimating the coefficients of multiple regression models in scenarios where the independent variables are highly correlated.,A regression analysis method that estimates the coefficients of multiple regression models in scenarios where the independent variables are highly correlated.,https://en.wikipedia.org/wiki/Ridge_regression|,,,,,AIO:RegressionAnalysis -AIO:LassoRegression,Machine Learning,AIO:MachineLearningSubset,,,Lasso Regression,0.00,A regression analysis method that performs both variable selection and regularization to enhance prediction accuracy and interpretability.,A regression analysis method that performs both variable selection and regularization to enhance prediction accuracy and interpretability.,https://en.wikipedia.org/wiki/Lasso_(statistics)|,,,,,AIO:RegressionAnalysis -AIO:SpatialRegression,Machine Learning,AIO:MachineLearningSubset,,,Spatial Regression,0.11,A regression method used to model spatial relationships.,A regression analysis method used to model spatial relationships.,https://gisgeography.com/spatial-regression-models-arcgis/|,,,,,AIO:RegressionAnalysis -AIO:LogisticRegression,Machine Learning,AIO:MachineLearningSubset,,,Logistic Regression,0.12,A statistical model that estimates the probability of an event occurring by modeling the log-odds of the event as a linear combination of one or more independent variables.,A regression analysis model that estimates the probability of an event occurring by modeling the log-odds of the event as a linear combination of one or more independent variables.,https://en.wikipedia.org/wiki/Logistic_regression|,,,,,AIO:RegressionAnalysis -AIO:LinearRegression,Machine Learning,AIO:MachineLearningSubset,,,Linear Regression,0.23,A linear approach for modeling the relationship between a scalar response and one or more explanatory variables.,A regression analysis model that is a linear approach for modeling the relationship between a scalar response and one or more explanatory variables.,https://en.wikipedia.org/wiki/Linear_regression|,,,,,AIO:RegressionAnalysis -AIO:LeastsquaresAnalysis,Machine Learning,AIO:MachineLearningSubset,,,Least-squares Analysis,,,A regression analysis which approximates the solution of overdetermined systems by minimizing the sum of the squares of the residuals.,https://en.wikipedia.org/wiki/Least_squares|,,,,,AIO:RegressionAnalysis -AIO:ProportionalHazardsModel,Machine Learning,AIO:MachineLearningSubset,,,Proportional Hazards Model,0.17,A survival modeling method where the unique effect of a unit increase in a covariate is multiplicative with respect to the hazard rate.,A regression analysis method for survival analysis where the unique effect of a unit increase in a covariate is multiplicative with respect to the hazard rate.,https://en.wikipedia.org/wiki/Proportional_hazards_model|,,,,,AIO:RegressionAnalysis|AIO:SurvivalAnalysis -AIO:GaussianNoiseLayer,Layer,AIO:LayerSubset,,,GaussianNoise Layer,0.89,"Apply additive zero-centered Gaussian noise. This is useful to mitigate overfitting (you could see it as a form of random data augmentation). Gaussian Noise (GS) is a natural choice as corruption process for real valued inputs. As it is a regularization layer, it is only active at training time.",A regularization layer that applies additive zero-centered Gaussian noise.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GaussianNoise,,,,,AIO:RegularizationLayer -AIO:AlphaDropoutLayer,Layer,AIO:LayerSubset,,,AlphaDropout Layer,0.74,"Applies Alpha Dropout to the input. Alpha Dropout is a Dropout that keeps mean and variance of inputs to their original values, in order to ensure the self-normalizing property even after this dropout. Alpha Dropout fits well to Scaled Exponential Linear Units by randomly setting activations to the negative saturation value.",A regularization layer that applies Alpha Dropout to the input keeping mean and variance of inputs to ensure self-normalizing property.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/AlphaDropout,,,,,AIO:RegularizationLayer -AIO:ActivityRegularizationLayer,Layer,AIO:LayerSubset,,,ActivityRegularization Layer,0.31,Layer that applies an update to the cost function based input activity.,A regularization layer that applies an update to the cost function based on input activity.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ActivityRegularization,,,,,AIO:RegularizationLayer -AIO:DropoutLayer,Layer,AIO:LayerSubset,,,Dropout Layer,0.92,"Applies Dropout to the input. The Dropout layer randomly sets input units to 0 with a frequency of rate at each step during training time, which helps prevent overfitting. Inputs not set to 0 are scaled up by 1/(1 - rate) such that the sum over all inputs is unchanged. Note that the Dropout layer only applies when training is set to True such that no values are dropped during inference. When using model.fit, training will be appropriately set to True automatically, and in other contexts, you can set the kwarg explicitly to True when calling the layer. (This is in contrast to setting trainable=False for a Dropout layer. trainable does not affect the layer's behavior, as Dropout does not have any variables/weights that can be frozen during training.)",A regularization layer that applies Dropout to the input,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Dropout,,,,,AIO:RegularizationLayer -AIO:GaussianDropoutLayer,Layer,AIO:LayerSubset,,,GaussianDropout Layer,0.75,"Apply multiplicative 1-centered Gaussian noise. As it is a regularization layer, it is only active at training time.",A regularization layer that applies multiplicative 1-centered Gaussian noise.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GaussianDropout,,,,,AIO:RegularizationLayer -AIO:SpatialDropout1DLayer,Layer,AIO:LayerSubset,,,SpatialDropout1D Layer,0.80,"Spatial 1D version of Dropout. This version performs the same function as Dropout, however, it drops entire 1D feature maps instead of individual elements. If adjacent frames within feature maps are strongly correlated (as is normally the case in early convolution layers) then regular dropout will not regularize the activations and will otherwise just result in an effective Learning rate decrease. In this case, SpatialDropout1D will help promote independence between feature maps and should be used instead.",A regularization layer that performs the same function as Dropout but drops entire 1D feature maps instead of individual elements.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/SpatialDropout1D,,,,,AIO:RegularizationLayer -AIO:SpatialDropout2DLayer,Layer,AIO:LayerSubset,,,SpatialDropout2D Layer,0.80,"Spatial 2D version of Dropout. This version performs the same function as Dropout, however, it drops entire 2D feature maps instead of individual elements. If adjacent pixels within feature maps are strongly correlated (as is normally the case in early convolution layers) then regular dropout will not regularize the activations and will otherwise just result in an effective Learning rate decrease. In this case, SpatialDropout2D will help promote independence between feature maps and should be used instead.a",A regularization layer that performs the same function as Dropout but drops entire 2D feature maps instead of individual elements.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/SpatialDropout2D,,,,,AIO:RegularizationLayer -AIO:SpatialDropout3DLayer,Layer,AIO:LayerSubset,,,SpatialDropout3D Layer,0.80,"Spatial 3D version of Dropout. This version performs the same function as Dropout, however, it drops entire 3D feature maps instead of individual elements. If adjacent voxels within feature maps are strongly correlated (as is normally the case in early convolution layers) then regular dropout will not regularize the activations and will otherwise just result in an effective Learning rate decrease. In this case, SpatialDropout3D will help promote independence between feature maps and should be used instead.",A regularization layer that performs the same function as Dropout but drops entire 3D feature maps instead of individual elements.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/SpatialDropout3D,,,,,AIO:RegularizationLayer -AIO:SparseLearning,Network,AIO:NetworkSubset,Sparse dictionary Learning|Sparse coding,,Sparse Learning,0.39,Finding sparse representations of input data as a linear combination of basic elements and identifying those elements.,A representation learning network that finds sparse representations of input data as a linear combination of basic elements and identifies those elements.,https://en.wikipedia.org/wiki/Sparse_dictionary_learning,,,,,AIO:RepresentationLearning -AIO:Cropping1DLayer,Layer,AIO:LayerSubset,,,Cropping1D Layer,0.55,Cropping layer for 1D input (e.g. temporal sequence). It crops along the time dimension (axis 1).,A layer that crops along the time dimension (axis 1) for 1D input.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Cropping1D,,,,,AIO:ReshapingLayer -AIO:FlattenLayer,Layer,AIO:LayerSubset,,,Flatten Layer,0.92,Flattens the input. Does not affect the batch size.,A layer that flattens the input,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Flatten,,,,,AIO:ReshapingLayer -AIO:PermuteLayer,Layer,AIO:LayerSubset,,,Permute Layer,0.55,Permutes the dimensions of the input according to a given pattern. Useful e.g. connecting RNNs and convnets.,A layer that permutes the dimensions of the input according to a given pattern.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Permute,,,,,AIO:ReshapingLayer -AIO:RepeatVectorLayer,Layer,AIO:LayerSubset,,,RepeatVector Layer,0.56,Repeats the input n times.,A layer that repeats the input n times.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RepeatVector,,,,,AIO:ReshapingLayer -AIO:ReshapeLayer,Layer,AIO:LayerSubset,,,Reshape Layer,0.30,Layer that reshapes inputs into the given shape.,A layer that reshapes the inputs into the given shape.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Reshape,,,,,AIO:ReshapingLayer -AIO:UpSampling1DLayer,Layer,AIO:LayerSubset,,,UpSampling1D Layer,0.52,Upsampling layer for 1D inputs. Repeats each temporal step size times along the time axis.,A layer that upsamples the input by repeating each temporal step size times along the time axis.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/UpSampling1D,,,,,AIO:ReshapingLayer -AIO:ZeroPadding3DLayer,Layer,AIO:LayerSubset,,,ZeroPadding3D Layer,0.93,Zero-padding layer for 3D data (spatial or spatio-temporal).,A layer that zero-pads the input along the depth,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ZeroPadding3D,,,,,AIO:ReshapingLayer -AIO:ZeroPadding2DLayer,Layer,AIO:LayerSubset,,,ZeroPadding2D Layer,0.88,"Zero-padding layer for 2D input (e.g. picture). This layer can add rows and columns of zeros at the top, bottom, left and right side of an image tensor.",A layer that zero-pads the input along the height and width dimensions.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ZeroPadding2D,,,,,AIO:ReshapingLayer -AIO:ZeroPadding1DLayer,Layer,AIO:LayerSubset,,,ZeroPadding1D Layer,0.87,Zero-padding layer for 1D input (e.g. temporal sequence).,A layer that zero-pads the input along the time axis.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ZeroPadding1D,,,,,AIO:ReshapingLayer -AIO:EcologicalFallacyBias,Bias,AIO:BiasSubset,Ecological Fallacy,,Ecological Fallacy Bias,0.32,Bias occurring when an inference about an individual is made based on their group membership.,A selection and sampling bias occurring when an inference about an individual is made based on their group membership.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:SelectionAndSamplingBias -AIO:SimponsParadoxBias,Bias,AIO:BiasSubset,Simpson's Paradox,,Simpon's Paradox Bias,#N/A,A statistical phenomenon where the association between two variables changes when controlling for another variable.,#N/A,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:SelectionAndSamplingBias -AIO:DataGenerationBias,Bias,AIO:BiasSubset,,,Data Generation Bias,0.41,Bias from adding synthetic or redundant data samples to a dataset.,A selection and sampling bias arising from adding synthetic or redundant data samples to a dataset.,https://en.wikipedia.org/wiki/Selection_bias|GTP-4o with Seppala et al. 2017,,,,,AIO:SelectionAndSamplingBias -AIO:TemporalBias,Bias,AIO:BiasSubset,,,Temporal Bias,0.36,Bias arising from differences in populations and behaviors over time.,A selection and sampling bias arising from differences in populations and behaviors over time.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:SelectionAndSamplingBias -AIO:MeasurementBias,Bias,AIO:BiasSubset,,,Measurement Bias,0.33,"Bias arising when features and labels are proxies for desired quantities, potentially leading to differential performance.",A selection and sampling bias arising when features and labels are proxies for desired quantities potentially leading to differential performance.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:SelectionAndSamplingBias -AIO:EvaluationBias,Bias,AIO:BiasSubset,,,Evaluation Bias,0.29,Bias arising when testing populations do not equally represent user populations or when inappropriate performance metrics are used.,A selection and sampling bias arising when testing populations do not equally represent user populations or when inappropriate performance metrics are used.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:SelectionAndSamplingBias -AIO:DetectionBias,Bias,AIO:BiasSubset,,,Detection Bias,0.46,"Systematic differences between groups in how outcomes are determined, potentially over- or underestimating effect size.",A selection and sampling bias characterized by systematic differences between groups in how outcomes are determined potentially over- or underestimating effect size.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:SelectionAndSamplingBias -AIO:PopulationBias,Bias,AIO:BiasSubset,,,Population Bias,0.36,Systematic distortions in demographics or other user characteristics between represented users and the target population.,A selection and sampling bias characterized by systematic distortions in demographics or other user characteristics between represented users and the target population.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:SelectionAndSamplingBias -AIO:RepresentationBias,Bias,AIO:BiasSubset,,,Representation Bias,0.41,"Bias due to non-random sampling of subgroups, making trends non-generalizable to new populations.",A selection and sampling bias due to non-random sampling of subgroups making trends non-generalizable to new populations.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:SelectionAndSamplingBias -AIO:UncertaintyBias,Bias,AIO:BiasSubset,,,Uncertainty Bias,0.42,"Bias favoring groups better represented in training data, due to less prediction uncertainty.",A selection and sampling bias favoring groups better represented in training data due to less prediction uncertainty.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:SelectionAndSamplingBias -AIO:ExclusionBias,Bias,AIO:BiasSubset,,,Exclusion Bias,0.28,Bias occurring when specific groups of user populations are excluded from testing and analysis.,A selection and sampling bias occurring when specific groups of user populations are excluded from testing and analysis.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:SelectionAndSamplingBias -AIO:PopularityBias,Bias,AIO:BiasSubset,,,Popularity Bias,0.44,"Selection bias where more popular items are more exposed, under-representing less popular items.",A selection and sampling bias where more popular items are more exposed under-representing less popular items.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:SelectionAndSamplingBias -AIO:AssociationRuleLearning,Machine Learning,AIO:MachineLearningSubset,,,Association Rule Learning,0.32,A rule-based machine learning method for discovering interesting relations between variables in large databases.,A supervised learning method focused on a rule-based approach for discovering interesting relations between variables in large databases.,https://en.wikipedia.org/wiki/Association_rule_learning|,,,,,AIO:SupervisedLearning -AIO:Classification,Machine Learning,AIO:MachineLearningSubset,,,Classification,0.44,"Methods that distinguish and distribute kinds of ""things"" into different groups.","A supervised learning task focused on methods that distinguish and distribute kinds of ""things"" into different groups.",https://en.wikipedia.org/wiki/Classification_(general_theory)|,,,,,AIO:SupervisedLearning -AIO:RegressionAnalysis,Machine Learning,AIO:MachineLearningSubset,Regression analysis|Regression model,,Regression Analysis,0.00,A set of statistical processes for estimating the relationships between a dependent variable and one or more independent variables.,A set of statistical processes for estimating the relationships between a dependent variable and one or more independent variables.,https://en.wikipedia.org/wiki/Regression_analysis|,,,,,AIO:SupervisedLearning|AIO:RegressionAnalysis -AIO:BoltzmannMachineNetwork,Network,AIO:NetworkSubset,BM|stochastic Hopfield network with hidden units|Sherrington–Kirkpatrick model with external field|stochastic Ising-Lenz-Little model,,Boltzmann Machine Network,0.57,"A Boltzmann machine is a type of stochastic recurrent neural network and Markov random field, translated from statistical physics for use in cognitive science.",A symmetrically connected network that is a type of stochastic recurrent neural network and Markov random field.,https://en.wikipedia.org/wiki/Boltzmann_machine,"Backfed Input, Probabilistic Hidden","Layers: Backfed Input, Probabilistic Hidden",,AIO:BackfedInputLayer|AIO:ProbabilisticHiddenLayer,AIO:SymmetricallyConnectedNetwork -AIO:HopfieldNetwork,Network,AIO:NetworkSubset,HN|Ising model of a neural network|Ising–Lenz–Little model,,Hopfield Network,0.64,"A Hopfield network is a type of recurrent artificial neural network that serves as a content-addressable (""associative"") memory system. It uses binary threshold nodes or continuous variables to store and recall memory patterns, providing a model for understanding human memory.",A symmetrically connected network that is a type of recurrent artificial neural network serving as a content-addressable memory system.,https://en.wikipedia.org/wiki/Hopfield_network,Backfed input,Layers: Backfed input,,AIO:BackfedInputLayer,AIO:SymmetricallyConnectedNetwork -AIO:TextVectorizationLayer,Layer,AIO:LayerSubset,,,TextVectorization Layer,0.18,A preprocessing layer which maps text features to integer sequences.,A preprocessing layer that maps text features to integer sequences.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/TextVectorization,,,,,AIO:TextPreprocessingLayer -AIO:KnowledgeTransfer,Preprocessing,AIO:PreprocessingSubset,Skill Acquisition|Inductive Transfer,Pretrained models|Adaptation,Knowledge Transfer,0.32,"The process by which knowledge is passed from one entity, such as a person, organization, or system, to another, facilitating learning and adaptation in the receiving entity through various methods such as teaching, training, or data exchange.",The process by which knowledge is passed from one entity such as a person organization or system to another facilitating learning and adaptation in the receiving entity through various methods such as teaching training or data exchange.,https://doi.org/10.1016/j.knosys.2015.01.010|,,,,,AIO:TrainingStrategies -AIO:CurriculumLearning,Preprocessing,AIO:PreprocessingSubset,Sequential Learning|Structured Learning,Sequential learning|Complexity grading,Curriculum Learning,0.13,"A training strategy in machine learning where models are trained on data in a meaningful order, starting with simpler examples and gradually increasing the complexity, to improve learning efficiency and model performance.",A training strategy in machine learning where models are trained on data in a meaningful order starting with simpler examples and gradually increasing the complexity to improve learning efficiency and model performance.,,,,,,AIO:TrainingStrategies -AIO:BidirectionalTransformerLanguageModel,Model,AIO:ModelSubset,Bidirectional Transformer LM|BERT,,Bidirectional Transformer Language Model,0.19,"A bidirectional transformer language model, such as BERT, uses the transformer architecture to build deep bidirectional representations by predicting masked tokens based on their context.",A transformer language model such as BERT that uses the transformer architecture to build deep bidirectional representations by predicting masked tokens based on their context.,https://arxiv.org/abs/1810.04805|https://en.wikipedia.org/wiki/BERT_(language_model)|,,,,,AIO:TransformerLanguageModel -AIO:TransformerLLM,Model,AIO:ModelSubset,Transformer Large Language Model,,Transformer LLM,0.15,"A transformer LLM is a neural network model with large training corpuses and large sets of parameters that uses the transformer architecture based on multi-head attention mechanisms, allowing it to contextualize tokens within a context window for effective language understanding and generation.",A transformer language model with large training corpuses and sets of parameters that uses the transformer architecture based on multi-head attention mechanisms allowing it to contextualize tokens within a context window for effective language understanding and generation.,https://en.wikipedia.org/wiki/Transformer_(deep_learning_architecture)|,,,,,AIO:TransformerLanguageModel -AIO:MultimodalTransformer,Model,AIO:ModelSubset,Multimodal Transformer,vision-language model|unified encoder,Multimodal Transformer,0.39,"A multimodal transformer processes and relates information from different modalities, such as text, images, and audio. It uses a shared embedding space and attention mechanism to learn joint representations across modalities.",A transformer network that processes and relates information from different modalities such as text images and audio using a shared embedding space and attention mechanism to learn joint representations across modalities.,,,,,,AIO:TransformerNetwork -AIO:DimensionalityReduction,Machine Learning,AIO:MachineLearningSubset,Dimension Reduction,,Dimensionality Reduction,0.29,The process of transforming data from a high-dimensional space into a lower-dimensional space while retaining meaningful properties of the original data.,A machine learning task focused on the process of transforming data from a high-dimensional space into a lower-dimensional space while retaining meaningful properties of the original data.,https://en.wikipedia.org/wiki/Dimensionality_reduction|,,,,,AIO:UnsupervisedLearning -AIO:AutoEncoderNetwork,Network,AIO:NetworkSubset,AE,,Auto Encoder Network,0.48,"An autoencoder is an artificial neural network used for learning efficient codings of unlabeled data, training the network to ignore insignificant data and regenerate input from encoding.",An unsupervised pretrained network that learns efficient codings of unlabeled data by training to ignore insignificant data and regenerate input from encoding.,https://en.wikipedia.org/wiki/Autoencoder,"Input, Hidden, Matched Output-Input","Layers: Input, Hidden, Matched Output-Input",,AIO:InputLayer|AIO:HiddenLayer|AIO:MatchedInputOutputLayer,AIO:UnsupervisedPretrainedNetwork -AIO:DeepBeliefNetwork,Network,AIO:NetworkSubset,DBN,,Deep Belief Network,0.57,"A deep belief network (DBN) is a generative graphical model composed of multiple layers of latent variables, learning to probabilistically reconstruct inputs and perform classification.",An unsupervised pretrained network composed of multiple layers of latent variables that learns to probabilistically reconstruct inputs and perform classification.,https://en.wikipedia.org/wiki/Deep_belief_network,"Backfed Input, Probabilistic Hidden, Hidden, Matched Output-Input","Layers: Backfed Input, Probabilistic Hidden, Hidden, Matched Output-Input",,AIO:BackfedInputLayer|AIO:ProbabilisticHiddenLayer|AIO:HiddenLayer|AIO:MatchedInputOutputLayer,AIO:UnsupervisedPretrainedNetwork -AIO:GenerativeAdversarialNetwork,Network,AIO:NetworkSubset,GAN,,Generative Adversarial Network,0.32,A generative adversarial network (GAN) is a machine learning framework where two neural networks contest in a game to generate new data with the same statistics as the training set.,An unsupervised pretrained network framework where two neural networks contest in a game to generate new data with the same statistics as the training set.,https://en.wikipedia.org/wiki/Generative_adversarial_network,"Backfed Input, Hidden, Matched Output-Input, Hidden, Matched Output-Input","Layers: Backfed Input, Hidden, Matched Output-Input, Hidden, Matched Output-Input",,AIO:BackfedInputLayer|AIO:HiddenLayer|AIO:MatchedInputOutputLayer,AIO:UnsupervisedPretrainedNetwork -AIO:ConceptDriftBias,Bias,AIO:BiasSubset,Concept Drift,,Concept Drift Bias,0.23,"Bias due to the use of a system outside its planned domain of application, causing performance gaps between laboratory settings and the real world.",A use and interpretation bias due to the use of a system outside its planned domain of application causing performance gaps between laboratory settings and the real world.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:UseAndInterpretationBias -AIO:DataDredgingBias,Bias,AIO:BiasSubset,Data Dredging,,Data Dredging Bias,0.22,Statistical bias where testing many hypotheses in a dataset may yield apparent statistical significance even when results are nonsignificant.,A use and interpretation bias where testing many hypotheses in a dataset may yield apparent statistical significance even when results are nonsignificant.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:UseAndInterpretationBias -AIO:ContentProductionBias,Bias,AIO:BiasSubset,,,Content Production Bias,0.63,"Bias from structural, lexical, semantic, and syntactic differences in user-generated content.",A use and interpretation bias arising from structural lexical semantic and syntactic differences in user-generated content.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:UseAndInterpretationBias -AIO:LinkingBias,Bias,AIO:BiasSubset,,,Linking Bias,0.45,"Bias arising when network attributes obtained from user connections, activities, or interactions misrepresent true user behavior.",A use and interpretation bias arising when network attributes obtained from user connections activities or interactions misrepresent true user behavior.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:UseAndInterpretationBias -AIO:FeedbackLoopBias,Bias,AIO:BiasSubset,,,Feedback Loop Bias,0.25,Effects occurring when an algorithm learns from user behavior and feeds that behavior back into the model.,A use and interpretation bias occurring when an algorithm learns from user behavior and feeds that behavior back into the model.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:UseAndInterpretationBias -AIO:ActivityBias,Bias,AIO:BiasSubset,,,Activity Bias,0.29,"Selection bias occurring when systems/platforms get training data from their most active users, rather than less active or inactive users.",A use and interpretation bias occurring when systems/platforms get training data from their most active users rather than less active or inactive users.,https://en.wikipedia.org/wiki/Interpretive_bias|GTP-4o with Seppala et al. 2017,,,,,AIO:UseAndInterpretationBias -AIO:EmergentBias,Bias,AIO:BiasSubset,,,Emergent Bias,0.24,Bias resulting from the use and reliance on algorithms across new or unanticipated contexts.,A use and interpretation bias resulting from the use and reliance on algorithms across new or unanticipated contexts.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,AIO:UseAndInterpretationBias -AIO:HostileAttributionBias,Bias,AIO:BiasSubset,,,Hostile Attribution Bias,0.40,Bias where individuals perceive benign or ambiguous behaviors as hostile.,A use and interpretation bias where individuals perceive benign or ambiguous behaviors as hostile.,https://en.wikipedia.org/wiki/Interpretive_bias|GTP-4o with Seppala et al. 2017,,,,,AIO:UseAndInterpretationBias -AIO:node2vecCBOW,Network,AIO:NetworkSubset,N2V-CBOW,CBOW,node2vec-CBOW,,,"A node2vec that predicts the current node from a window of surrounding context nodes, with the order of context nodes not influencing prediction.",https://en.wikipedia.org/wiki/Word2vec,"Input, Hidden, Output","Layers: Input, Hidden, Output",,AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:word2vec -AIO:word2vecCBOW,Network,AIO:NetworkSubset,W2V-CBOW,CBOW,word2vec-CBOW,,,"A word2vec that predicts the current word from a window of surrounding context words, ignoring the order of context words.",https://en.wikipedia.org/wiki/Word2vec,"Input, Hidden, Output","Layers: Input, Hidden, Output",,AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:word2vec -AIO:word2vecSkipGram,Network,AIO:NetworkSubset,W2V-SkipGram,SkipGram,word2vec-SkipGram,,,"A word2vec that predicts surrounding context words from the current word, giving more weight to nearby context words than distant ones.",https://en.wikipedia.org/wiki/Word2vec,"Input, Hidden, Output","Layers: Input, Hidden, Output",,AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:word2vec -AIO:Model,Model,AIO:ModelSubset,Model,,Model,0.38,"A model is an abstract representation of a complex system, generally assembled as a set of logical, mathematical, or conceptual properties to simulate or understand the system's behavior.",An abstract representation of a complex system generally assembled as a set of logical mathematical or conceptual properties to simulate or understand the system's behavior.,https://en.wikipedia.org/wiki/Mathematical_model|,,,,,owl:Thing -AIO:MachineLearning,,,,,Machine Learning,0.00,A field of inquiry devoted to understanding and building methods that learn from data to improve performance on a set of tasks.,A field of inquiry devoted to understanding and building methods that learn from data to improve performance on a set of tasks.,https://doi.org/10.6028/NIST.SP.1270|GTP-4o with Seppala et al. 2017,,,,,owl:Thing -AIO:Bias,Bias,AIO:BiasSubset,,,Bias,0.18,Systematic error introduced into sampling or testing by selecting or encouraging one outcome or answer over others.,A systematic error introduced into sampling or testing by selecting or encouraging one outcome or answer over others.,https://www.merriam-webster.com/dictionary/bias|GTP-4o with Seppala et al. 2017,,,,,owl:Thing -AIO:Function,Class,AIO:ClassSubset,,,Function,0.15,A mathematical rule that gives the value of a dependent variable corresponding to specified values of one or more independent variables.,A mathematical rule that gives the value of a dependent variable corresponding to specified values of independent variables.,https://www.sciencedirect.com/topics/mathematics/mathematical-function|,,,,,owl:Thing -AIO:Layer,Class,AIO:ClassSubset,,,Layer,0.00,A structure or network topology in a deep learning model that takes information from previous layers and passes it to the next layer.,A structure or network topology in a deep learning model that takes information from previous layers and passes it to the next layer.,https://en.wikipedia.org/wiki/Layer_(deep_learning),,,,,owl:Thing -AIO:Network,Class,AIO:ClassSubset,,,Network,0.29,"A system of interconnected nodes or entities for communication, computation, or data exchange.",A system of interconnected nodes or entities for communication computation or data exchange.,,,,,,owl:Thing -AIO:Preprocessing,Preprocessing,AIO:PreprocessingSubset,,,Preprocessing,0.26,"The series of steps applied to raw data before it is used in a machine learning model, including tasks such as normalization, scaling, encoding, and transformation, to ensure the data is in an appropriate format and quality for analysis.",The series of steps applied to raw data before it is used in a machine learning model including tasks such as normalization scaling encoding and transformation to ensure the data is in an appropriate format and quality for analysis.,https://doi.org/10.1109/ICDE.2019.00245|,,,,,owl:Thing \ No newline at end of file +ID,Category,Synomyms,Related synonyms,Name,ExactSynonym and label identical,description jaccard diff,Legacy Description,description details comment,Claude opus Aristotelian definition,Curated Claude opus Aristotelian definition,Reference (DOI),Layers comment,layer parts,Parent +ID,AI oio:inSubset SPLIT=|,A oio:hasExactSynonym SPLIT=|,A oio:hasRelatedSynonym SPLIT=|,LABEL,,,,A rdfs:comment,,A IAO:0000115,>AI oio:hasDbXref SPLIT=|,A rdfs:comment,SC BFO:0000051 some % SPLIT=|,SC % SPLIT=| +AIO:SoftmaxFunction,AIO:FunctionSubset|AIO:ActivationFunctionSubset,,,Softmax Function,FALSE,0.70,"The elements of the output vector are in range (0, 1) and sum to 1. Each vector is handled independently. The axis argument sets which axis of the input the function is applied along. Softmax is often used as the activation for the last layer of a classification network because the result could be interpreted as a probability distribution. The softmax of each vector x is computed as exp(x) / tf.reduce_sum(exp(x)). The input values in are the log-odds of the resulting probability.","The elements of the output vector are in range (0, 1) and sum to 1. Each vector is handled independently. The axis argument sets which axis of the input the function is applied along. Softmax is often used as the activation for the last layer of a classification network because the result could be interpreted as a probability distribution. The softmax of each vector x is computed as exp(x) / tf.reduce_sum(exp(x)). The input values in are the log-odds of the resulting probability.",An activation function where the elements of the output vector are in range (0 1) and sum to 1 and each vector is handled independently.,A mathematical function where the elements of the output vector are in range (0 1) and sum to 1 and each vector is handled independently.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/softmax,,,AIO:MathematicalFunction +AIO:SigmoidFunction,AIO:FunctionSubset|AIO:ActivationFunctionSubset,tore,,Sigmoid Function,FALSE,0.66,"Applies the sigmoid activation function sigmoid(x) = 1 / (1 + exp(-x)). For small values (<-5), sigmoid returns a value close to zero, and for large values (>5) the result of the function gets close to 1. Sigmoid is equivalent to a 2-element Softmax, where the second element is assumed to be zero. The sigmoid function always returns a value between 0 and 1.","Applies the sigmoid activation function sigmoid(x) = 1 / (1 + exp(-x)). For small values (<-5), sigmoid returns a value close to zero, and for large values (>5) the result of the function gets close to 1. Sigmoid is equivalent to a 2-element Softmax, where the second element is assumed to be zero. The sigmoid function always returns a value between 0 and 1.",An activation function that applies the sigmoid activation function sigmoid(x) = 1 / (1 + exp(-x)) always returning a value between 0 and 1.,A mathematical function that applies the sigmoid activation function sigmoid(x) = 1 / (1 + exp(-x)) always returning a value between 0 and 1.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/sigmoid,,,AIO:MathematicalFunction +AIO:LinearFunction,AIO:FunctionSubset|AIO:ActivationFunctionSubset,,,Linear Function,FALSE,0.36,A linear function has the form f(x) = a + bx.,,An activation function that has the form f(x) = a + bx.,A mathematical function that has the form f(x) = a + bx.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/linear,,,AIO:MathematicalFunction +AIO:SoftplusFunction,AIO:FunctionSubset|AIO:ActivationFunctionSubset,,,Softplus Function,FALSE,0.64,softplus(x) = log(exp(x) + 1),softplus(x) = log(exp(x) + 1),An activation function that is softplus(x) = log(exp(x) + 1).,A mathematical function that is softplus(x) = log(exp(x) + 1).,https://www.tensorflow.org/api_docs/python/tf/keras/activations/softplus,,,AIO:MathematicalFunction +AIO:SoftsignFunction,AIO:FunctionSubset|AIO:ActivationFunctionSubset,,,Softsign Function,FALSE,0.54,softsign(x) = x / (abs(x) + 1),softsign(x) = x / (abs(x) + 1),An activation function that is softsign(x) = x / (abs(x) + 1).,A mathematical function that is softsign(x) = x / (abs(x) + 1).,https://www.tensorflow.org/api_docs/python/tf/keras/activations/softsign,,,AIO:MathematicalFunction +AIO:TanhFunction,AIO:FunctionSubset|AIO:ActivationFunctionSubset,hyperbolic tangent,,Tanh Function,FALSE,0.70,Hyperbolic tangent activation function.,Hyperbolic tangent activation function.,An activation function that is the hyperbolic tangent activation function.,A mathematical function that is the hyperbolic tangent activation function.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/tanh,,,AIO:MathematicalFunction +AIO:ELUFunction,AIO:FunctionSubset|AIO:ActivationFunctionSubset,Exponential Linear Unit|ELU,,ELU Function,FALSE,0.68,The exponential linear unit (ELU) with alpha > 0 is: x if x > 0 and alpha * (exp(x) - 1) if x < 0 The ELU hyperparameter alpha controls the value to which an ELU saturates for negative net inputs. ELUs diminish the vanishing gradient effect. ELUs have negative values which pushes the mean of the activations closer to zero. Mean activations that are closer to zero enable faster Learning as they bring the gradient closer to the natural gradient. ELUs saturate to a negative value when the argument gets smaller. Saturation means a small derivative which decreases the variation and the information that is propagated to the next layer.,The exponential linear unit (ELU) with alpha > 0 is: x if x > 0 and alpha * (exp(x) - 1) if x < 0 The ELU hyperparameter alpha controls the value to which an ELU saturates for negative net inputs. ELUs diminish the vanishing gradient effect. ELUs have negative values which pushes the mean of the activations closer to zero. Mean activations that are closer to zero enable faster Learning as they bring the gradient closer to the natural gradient. ELUs saturate to a negative value when the argument gets smaller. Saturation means a small derivative which decreases the variation and the information that is propagated to the next layer.,An activation function that is x if x > 0 and alpha * (exp(x) - 1) if x < 0 where alpha controls the value to which an ELU saturates for negative net inputs.,A mathematical function that is x if x > 0 and alpha * (exp(x) - 1) if x < 0 where alpha controls the value to which an ELU saturates for negative net inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/elu,,,AIO:MathematicalFunction +AIO:SwishFunction,AIO:FunctionSubset|AIO:ActivationFunctionSubset,,,Swish Function,FALSE,0.56,"x*sigmoid(x). It is a smooth, non-monotonic function that consistently matches or outperforms ReLU on deep networks, it is unbounded above and bounded below.","x*sigmoid(x). It is a smooth, non-monotonic function that consistently matches or outperforms ReLU on deep networks, it is unbounded above and bounded below.",An activation function that is x*sigmoid(x) a smooth non-monotonic function that consistently matches or outperforms ReLU on deep networks.,A mathematical function that is x*sigmoid(x) a smooth non-monotonic function that consistently matches or outperforms ReLU on deep networks.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/swish,,,AIO:MathematicalFunction +AIO:SELUFunction,AIO:FunctionSubset|AIO:ActivationFunctionSubset,Scaled Exponential Linear Unit|SELU,,SELU Function,FALSE,0.16,The SELU activation function multiplies scale (> 1) with the output of the ELU function to ensure a slope larger than one for positive inputs.,,An activation function that multiplies scale (> 1) with the output of the ELU function to ensure a slope larger than one for positive inputs.,A mathematical function that multiplies scale (> 1) with the output of the ELU function to ensure a slope larger than one for positive inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/selu,,,AIO:MathematicalFunction +AIO:ReLUFunction,AIO:FunctionSubset|AIO:ActivationFunctionSubset,Rectified Linear Unit|ReLU,,ReLU Function,FALSE,0.50,"The ReLU activation function returns: max(x, 0), the element-wise maximum of 0 and the input tensor.","The ReLU activation function returns: max(x, 0), the element-wise maximum of 0 and the input tensor.",An activation function that returns max(x 0) the element-wise maximum of 0 and the input tensor.,A mathematical function that returns max(x 0) the element-wise maximum of 0 and the input tensor.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/relu,,,AIO:MathematicalFunction +AIO:LazyInstanceNorm1DLayer,AIO:LayerSubset|AIO:InstanceNormalizationLayerSubset,LazyInstanceNorm1D|LazyInstanceNorm1D|LazyInstanceNorm1D,,LazyInstanceNorm1D Layer,FALSE,0.81,A torch.nn.InstanceNorm1D module with lazy initialization of the num_features argument of the InstanceNorm1D that is inferred from the input.size(1).,A torch.nn.InstanceNorm1D module with lazy initialization of the num_features argument of the InstanceNorm1D that is inferred from the input.size(1).,An instance normalization layer that lazily initializes the num_features argument from the input size for 1D data.,A normalization layer that lazily initializes the num_features argument from the input size for 1D data.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:NormalizationLayer +AIO:LazyInstanceNorm2DLayer,AIO:LayerSubset|AIO:InstanceNormalizationLayerSubset,LazyInstanceNorm2D|LazyInstanceNorm2D|LazyInstanceNorm2D,,LazyInstanceNorm2D Layer,FALSE,0.81,A torch.nn.InstanceNorm2D module with lazy initialization of the num_features argument of the InstanceNorm2D that is inferred from the input.size(1).,A torch.nn.InstanceNorm2D module with lazy initialization of the num_features argument of the InstanceNorm2D that is inferred from the input.size(1).,An instance normalization layer that lazily initializes the num_features argument from the input size for 2D data.,A normalization layer that lazily initializes the num_features argument from the input size for 2D data.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:NormalizationLayer +AIO:LazyInstanceNorm3DLayer,AIO:LayerSubset|AIO:InstanceNormalizationLayerSubset,LazyInstanceNorm3D|LazyInstanceNorm3D|LazyInstanceNorm3D,,LazyInstanceNorm3D Layer,FALSE,0.81,A torch.nn.InstanceNorm3D module with lazy initialization of the num_features argument of the InstanceNorm3D that is inferred from the input.size(1).,A torch.nn.InstanceNorm3D module with lazy initialization of the num_features argument of the InstanceNorm3D that is inferred from the input.size(1).,An instance normalization layer that lazily initializes the num_features argument from the input size for 3D data.,A normalization that lazily initializes the num_features argument from the input size for 3D data.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:NormalizationLayer +AIO:BatchNorm1DLayer,AIO:LayerSubset,BatchNorm1D|BatchNorm1D|BatchNorm1D,,BatchNorm1D Layer,FALSE,0.78,Applies Batch Normalization over a 2D or 3D input as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift .,Applies Batch Normalization over a 2D or 3D input as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift .,A batch normalization layer that applies Batch Normalization over a 2D or 3D input.,A batch normalization layer that applies Batch Normalization over a 2D or 3D input.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:BatchNormalizationLayer +AIO:BatchNorm2DLayer,AIO:LayerSubset,BatchNorm2D|BatchNorm2D|BatchNorm2D,,BatchNorm2D Layer,FALSE,0.87,Applies Batch Normalization over a 4D input (a mini-batch of 2D inputs with additional channel dimension) as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift .,Applies Batch Normalization over a 4D input (a mini-batch of 2D inputs with additional channel dimension) as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift .,A batch normalization layer that applies Batch Normalization over a 4D input.,A batch normalization layer that applies Batch Normalization over a 4D input.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:BatchNormalizationLayer +AIO:BatchNorm3DLayer,AIO:LayerSubset,BatchNorm3D|BatchNorm3D|BatchNorm3D,,BatchNorm3D Layer,FALSE,0.87,Applies Batch Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension) as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift .,Applies Batch Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension) as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift .,A batch normalization layer that applies Batch Normalization over a 5D input.,A batch normalization layer that applies Batch Normalization over a 5D input.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:BatchNormalizationLayer +AIO:SyncBatchNormLayer,AIO:LayerSubset,SyncBatchNorm|SyncBatchNorm,,SyncBatchNorm Layer,FALSE,0.95,Applies Batch Normalization over a N-Dimensional input (a mini-batch of [N-2]D inputs with additional channel dimension) as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift .,Applies Batch Normalization over a N-Dimensional input (a mini-batch of [N-2]D inputs with additional channel dimension) as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift .,A batch normalization layer that applies synchronous Batch Normalization across multiple devices.,A batch normalization layer that applies synchronous Batch Normalization across multiple devices.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:BatchNormalizationLayer +AIO:LazyBatchNorm1DLayer,AIO:LayerSubset,LazyBatchNorm1D|LazyBatchNorm1D|LazyBatchNorm1D,,LazyBatchNorm1D Layer,FALSE,0.77,A torch.nn.BatchNorm1D module with lazy initialization of the num_features argument of the BatchNorm1D that is inferred from the input.size(1).,A torch.nn.BatchNorm1D module with lazy initialization of the num_features argument of the BatchNorm1D that is inferred from the input.size(1).,A batch normalization layer that lazily initializes the num_features argument from the input size for 1D data.,A batch normalization layer that lazily initializes the num_features argument from the input size for 1D data.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:BatchNormalizationLayer +AIO:LazyBatchNorm2DLayer,AIO:LayerSubset,LazyBatchNorm2D|LazyBatchNorm2D|LazyBatchNorm2D,,LazyBatchNorm2D Layer,FALSE,0.77,A torch.nn.BatchNorm2D module with lazy initialization of the num_features argument of the BatchNorm2D that is inferred from the input.size(1).,A torch.nn.BatchNorm2D module with lazy initialization of the num_features argument of the BatchNorm2D that is inferred from the input.size(1).,A batch normalization layer that lazily initializes the num_features argument from the input size for 2D data.,A batch normalization layer that lazily initializes the num_features argument from the input size for 2D data.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:BatchNormalizationLayer +AIO:LazyBatchNorm3DLayer,AIO:LayerSubset,LazyBatchNorm3D|LazyBatchNorm3D|LazyBatchNorm3D,,LazyBatchNorm3D Layer,FALSE,0.77,A torch.nn.BatchNorm3D module with lazy initialization of the num_features argument of the BatchNorm3D that is inferred from the input.size(1).,A torch.nn.BatchNorm3D module with lazy initialization of the num_features argument of the BatchNorm3D that is inferred from the input.size(1).,A batch normalization layer that lazily initializes the num_features argument from the input size for 3D data.,A batch normalization layer that lazily initializes the num_features argument from the input size for 3D data.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:BatchNormalizationLayer +AIO:ComputationalBias,AIO:BiasSubset,Statistical Bias,,Computational Bias,FALSE,0.29,,,A bias caused by differences between results and facts in the process of data analysis (including the source of data the estimator chose) and analysis methods.,A bias caused by differences between results and facts in the process of data analysis (including the source of data the estimator chose) and analysis methods.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:Bias +AIO:IndividualBias,AIO:BiasSubset,,,Individual Bias,FALSE,0.32,"A persistent point of view or limited list of such points of view applied by an individual, such as ""parent,"" ""academic,"" or ""professional.""",,A persistent point of view or limited list of such points of view applied by an individual.,"A bias characterized by a persistent point of view or limited list of such points of view, applied by an individual.",https://develop.consumerium.org/wiki/Individual_bias,,,AIO:Bias +AIO:SocietalBias,AIO:BiasSubset,,,Societal Bias,FALSE,0.33,"Bias characterized by being for or against groups or individuals based on social identities, demographic factors, or immutable physical characteristics, often manifesting as stereotypes.",,A bias characterized by being for or against groups or individuals based on social identities demographic factors or immutable physical characteristics often manifesting as stereotypes.,A systemic bias characterized by being for or against groups or individuals based on social identities demographic factors or immutable physical characteristics often manifesting as stereotypes.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:SystemicBias +AIO:HistoricalBias,AIO:BiasSubset,,,Historical Bias,FALSE,0.69,"Long-standing biases encoded in society over time, distinct from biases in historical description or the interpretation of history, such as viewing the larger world from a Western or European perspective.","Long-standing biases encoded in society over time, distinct from biases in historical description or the interpretation of history, such as viewing the larger world from a Western or European perspective.",A bias characterized by long-standing biases encoded in society over time distinct from biases in historical description or interpretation.,A bias characterized by long-standing biases encoded in society over time distinct from biases in historical description or interpretation.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:Bias +AIO:SunkCostFallacyBias,AIO:BiasSubset,Sunk Cost Fallacy,,Sunk Cost Fallacy Bias,FALSE,0.40,"The tendency to continue an endeavor due to previously invested resources, despite costs outweighing benefits.","The tendency to continue an endeavor due to previously invested resources, despite costs outweighing benefits.",A bias characterized by the tendency to continue an endeavor due to previously invested resources despite costs outweighing benefits.,A bias characterized by the tendency to continue an endeavor due to previously invested resources despite costs outweighing benefits.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:Bias +AIO:InstitutionalBias,AIO:BiasSubset,,,Institutional Bias,FALSE,0.43,"Bias exhibited at the level of entire institutions, where practices or norms result in the favoring or disadvantaging of certain social groups, such as institutional racism or sexism.","Bias exhibited at the level of entire institutions, where practices or norms result in the favoring or disadvantaging of certain social groups, such as institutional racism or sexism.",A bias exhibited at the level of entire institutions where practices or norms result in the favoring or disadvantaging of certain social groups.,A systemic bias exhibited at the level of entire institutions where practices or norms result in the favoring or disadvantaging of certain social groups.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:SystemicBias +AIO:HumanBias,AIO:BiasSubset,,,Human Bias,FALSE,0.39,"Systematic errors in human thought based on heuristic principles, leading to simplified judgmental operations.",,A systematic error in human thought based on heuristic principles leading to simplified judgmental operations.,A bias in human thought based on heuristic principles leading to simplified judgmental operations.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:Bias +AIO:SystemicBias,AIO:BiasSubset,,,Systemic Bias,FALSE,0.15,Biases resulting from procedures and practices of particular institutions that operate in ways which result in certain social groups being advantaged or favored and others being disadvantaged or devalued.,,A bias resulting from procedures and practices of institutions that operate in ways which result in certain social groups being advantaged or favored and others being disadvantaged or devalued.,A bias resulting from procedures and practices of systems that operate in ways which result in certain social groups being advantaged or favored and others being disadvantaged or devalued.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:Bias +AIO:SupervisedBiclustering,AIO:MachineLearningSubset,Supervised Block Clustering|Supervised Co-clustering|Supervised Two-mode Clustering|Supervised Two-way Clustering|Supervised Joint Clustering,,Supervised Biclustering,FALSE,0.36,"Methods that simultaneously cluster the rows and columns of a labeled matrix, considering data labels to enhance cluster coherence.",,A biclustering task focused on methods that simultaneously cluster the rows and columns of a labeled matrix considering data labels to enhance cluster coherence.,A biclustering task focused on methods that simultaneously cluster the rows and columns of a labeled matrix considering data labels to enhance cluster coherence.,https://en.wikipedia.org/wiki/Biclustering,,,AIO:Biclustering +AIO:UnsupervisedBiclustering,AIO:MachineLearningSubset,Unsupervised Block Clustering|Unsupervised Co-clustering|Unsupervised Two-mode Clustering|Unsupervised Two-way Clustering|Unsupervised Joint Clustering,,Unsupervised Biclustering,FALSE,0.28,Methods that simultaneously cluster the rows and columns of an unlabeled input matrix to identify submatrices with coherent patterns.,,A biclustering task focused on methods that simultaneously cluster the rows and columns of an unlabeled input matrix to identify submatrices with coherent patterns.,A biclustering task focused on methods that simultaneously cluster the rows and columns of an unlabeled input matrix to identify submatrices with coherent patterns.,https://en.wikipedia.org/wiki/Biclustering,,,AIO:Biclustering +AIO:RestrictedBoltzmannMachine,AIO:NetworkSubset,RBM,,Restricted Boltzmann Machine,FALSE,0.35,A restricted Boltzmann machine (RBM) is a generative stochastic neural network that learns the probability distribution of its input data.,,A Boltzmann machine network that learns the probability distribution of its input data.,A Boltzmann machine network that learns the probability distribution of its input data.,https://en.wikipedia.org/wiki/Restricted_Boltzmann_machine,"Layers: Backfed Input, Probabilistic Hidden",AIO:BackfedInputLayer|AIO:ProbabilisticHiddenLayer,AIO:BoltzmannMachineNetwork +AIO:CategoryEncodingLayer,AIO:LayerSubset,,,CategoryEncoding Layer,FALSE,0.78,"A preprocessing layer which encodes integer features. This layer provides options for condensing data into a categorical encoding when the total number of tokens are known in advance. It accepts integer values as inputs, and it outputs a dense or sparse representation of those inputs. For integer inputs where the total number of tokens is not known, use tf.keras.layers.IntegerLookup instead.","A preprocessing layer which encodes integer features. This layer provides options for condensing data into a categorical encoding when the total number of tokens are known in advance. It accepts integer values as inputs, and it outputs a dense or sparse representation of those inputs. For integer inputs where the total number of tokens is not known, use tf.keras.layers.IntegerLookup instead.",A categorical features preprocessing layer that encodes integer features providing options for condensing data into a categorical encoding.,A categorical features preprocessing layer that encodes integer features providing options for condensing data into a categorical encoding.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/CategoryEncoding,,,AIO:CategoricalFeaturesPreprocessingLayer +AIO:IntegerLookupLayer,AIO:LayerSubset,,,IntegerLookup Layer,FALSE,0.25,A preprocessing layer which maps integer features to contiguous ranges.,,A categorical features preprocessing layer that maps integer features to contiguous ranges.,A categorical features preprocessing layer that maps integer features to contiguous ranges.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/IntegerLookup,,,AIO:CategoricalFeaturesPreprocessingLayer +AIO:StringLookupLayer,AIO:LayerSubset,,,StringLookup Layer,FALSE,0.25,A preprocessing layer which maps string features to integer indices.,,A categorical features preprocessing layer that maps string features to integer indices.,A categorical features preprocessing layer that maps string features to integer indices.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/StringLookup,,,AIO:CategoricalFeaturesPreprocessingLayer +AIO:HashingLayer,AIO:LayerSubset,,,Hashing Layer,FALSE,0.89,"A preprocessing layer which hashes and bins categorical features. This layer transforms categorical inputs to hashed output. It element-wise converts a ints or strings to ints in a fixed range. The stable hash function uses tensorflow::ops::Fingerprint to produce the same output consistently across all platforms. This layer uses FarmHash64 by default, which provides a consistent hashed output across different platforms and is stable across invocations, regardless of device and context, by mixing the input bits thoroughly. If you want to obfuscate the hashed output, you can also pass a random salt argument in the constructor. In that case, the layer will use the SipHash64 hash function, with the salt value serving as additional input to the hash function.","A preprocessing layer which hashes and bins categorical features. This layer transforms categorical inputs to hashed output. It element-wise converts a ints or strings to ints in a fixed range. The stable hash function uses tensorflow::ops::Fingerprint to produce the same output consistently across all platforms. This layer uses FarmHash64 by default, which provides a consistent hashed output across different platforms and is stable across invocations, regardless of device and context, by mixing the input bits thoroughly. If you want to obfuscate the hashed output, you can also pass a random salt argument in the constructor. In that case, the layer will use the SipHash64 hash function, with the salt value serving as additional input to the hash function.",A categorical features preprocessing layer which hashes and bins categorical features.,A categorical features preprocessing layer which hashes and bins categorical features.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Hashing,,,AIO:CategoricalFeaturesPreprocessingLayer +AIO:KnearestNeighborClassificationAlgorithm,AIO:MachineLearningSubset,KNN Classification|K-NN Classification,,K-nearest Neighbor Classification Algorithm,FALSE,#N/A,,,"A classification and clustering that classifies objects by a plurality vote of its neighbors, assigning each object to the class most common among its k nearest neighbors.","A classification and clustering that classifies objects by a plurality vote of its neighbors, assigning each object to the class most common among its k nearest neighbors.",https://en.wikipedia.org/wiki/K-nearest_neighbors_algorithm,,,AIO:Clustering +AIO:BinaryClassification,AIO:MachineLearningSubset,,,Binary Classification,FALSE,0.39,Methods that classify elements into two groups based on a classification rule.,,A machine learning task focused on methods that classify elements into two groups based on a classification rule.,A classification focused on methods that classify elements into two groups based on a classification rule.,https://en.wikipedia.org/wiki/Binary_classification,,,AIO:Classification +AIO:MulticlassClassification,AIO:MachineLearningSubset,Multinomial Classification,,Multiclass Classification,FALSE,0.44,Methods that classify instances into one of three or more classes.,Methods that classify instances into one of three or more classes.,A machine learning task focused on methods that classify instances into one of three or more classes.,A classification focused on methods that classify instances into one of three or more classes.,https://en.wikipedia.org/wiki/Multiclass_classification,,,AIO:Classification +AIO:HierarchicalClassification,AIO:MachineLearningSubset,,,Hierarchical Classification,FALSE,0.50,Methods that group things according to a hierarchy.,Methods that group things according to a hierarchy.,A classification task focused on methods that group things according to a hierarchy.,A classification focused on methods that group things according to a hierarchy.,https://en.wikipedia.org/wiki/Hierarchical_classification,,,AIO:Classification +AIO:DecisionTree,AIO:MachineLearningSubset,,,Decision Tree,FALSE,0.41,"A decision support tool that uses a tree-like model of decisions and their possible consequences, including chance event outcomes, resource costs, and utilities.","A decision support tool that uses a tree-like model of decisions and their possible consequences, including chance event outcomes, resource costs, and utilities.",A machine learning model that uses a tree-like model of decisions and their possible consequences including chance event outcomes resource costs and utilities.,A classification that uses a tree-like model of decisions and their possible consequences including chance event outcomes resource costs and utilities.,https://en.wikipedia.org/wiki/Decision_tree,,,AIO:Classification +AIO:UnsupervisedClustering,AIO:MachineLearningSubset,Unsupervised Cluster Analysis,,Unsupervised Clustering,FALSE,0.26,Methods that group a set of unlabeled objects such that objects in the same group are more similar to each other than to those in other groups.,,A clustering task focused on methods that group a set of unlabeled objects such that objects in the same group are more similar to each other than to those in other groups.,A clustering focused on methods that group a set of unlabeled objects such that objects in the same group are more similar to each other than to those in other groups.,https://en.wikipedia.org/wiki/Cluster_analysis,,,AIO:Clustering +AIO:SupervisedClustering,AIO:MachineLearningSubset,Supervised Cluster Analysis,,Supervised Clustering,FALSE,0.38,"Methods that group labeled objects such that objects in the same group have similar labels, relative to those in other groups.",,A clustering task focused on methods that group labeled objects such that objects in the same group have similar labels relative to those in other groups.,A clustering focused on methods that group labeled objects such that objects in the same group have similar labels relative to those in other groups.,https://en.wikipedia.org/wiki/Cluster_analysis,,,AIO:Clustering +AIO:HierarchicalClustering,AIO:MachineLearningSubset,HCL,,Hierarchical Clustering,FALSE,0.55,Methods that build a hierarchy of clusters.,Methods that build a hierarchy of clusters.,A clustering method that builds a hierarchy of clusters.,A clustering that builds a hierarchy of clusters.,https://en.wikipedia.org/wiki/Hierarchical_clustering,,,AIO:Clustering +AIO:DunningKrugerEffectBias,AIO:BiasSubset,Dunning-Kruger Effect,,Dunning-Kruger Effect Bias,FALSE,#N/A,,,A cognitive bias in which people with low ability in an area overestimate that ability. Often measured by comparing self-assessment with objective performance.,A cognitive bias in which people with low ability in an area overestimate that ability. Often measured by comparing self-assessment with objective performance.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:CognitiveBias +AIO:UseAndInterpretationBias,AIO:BiasSubset,Interpretive Bias,,Use And Interpretation Bias,FALSE,0.67,"Bias inappropriately analyzing ambiguous stimuli, scenarios, and events.","Bias inappropriately analyzing ambiguous stimuli, scenarios, and events.",A computational bias characterized by inappropriately analyzing ambiguous stimuli scenarios and events.,A computational bias characterized by inappropriately analyzing ambiguous stimuli scenarios and events.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:ComputationalBias +AIO:SelectionAndSamplingBias,AIO:BiasSubset,Selection Bias|Sampling Bias|Selection Effect,,Selection And Sampling Bias,FALSE,0.50,"Bias introduced by non-random selection of individuals, groups, or data, failing to ensure representativeness.","Bias introduced by non-random selection of individuals, groups, or data, failing to ensure representativeness.",A computational bias introduced by non-random selection of individuals groups or data failing to ensure representativeness.,A computational bias introduced by non-random selection of individuals groups or data failing to ensure representativeness.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:ComputationalBias +AIO:ProcessingBias,AIO:BiasSubset,Validation Bias,,Processing Bias,FALSE,0.59,"Judgment modulated by affect, influenced by the level of efficacy and efficiency in information processing; often referred to as aesthetic judgment in cognitive sciences.","Judgment modulated by affect, influenced by the level of efficacy and efficiency in information processing; often referred to as aesthetic judgment in cognitive sciences.",A computational bias resulting from judgment modulated by affect influenced by the level of efficacy and efficiency in information processing.,A computational bias resulting from judgment modulated by affect influenced by the level of efficacy and efficiency in information processing.,https://en.wikipedia.org/wiki/Bias_(statistics),,,AIO:ComputationalBias +AIO:Convolution1DLayer,AIO:LayerSubset,Conv1D Layer|nn.Conv1D|Conv1D|Convolution1D|Convolution1D,,Convolution1D Layer,FALSE,0.33,1D convolution layer (e.g. temporal convolution).,,A layer that implements 1D convolution (e.g. temporal convolution).,A convolutional layer that implements 1D convolution (e.g. temporal convolution).,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv1D,,,AIO:ConvolutionalLayer +AIO:Convolution2DLayer,AIO:LayerSubset,Conv2D Layer|nn.Conv2D|Conv2D|Convolution2D|Convolution2D,,Convolution2D Layer,FALSE,0.90,"2D convolution layer (e.g. spatial convolution over images). This layer creates a convolution kernel that is convolved with the layer input to produce a tensor of outputs. If use_bias is True, a bias vector is created and added to the outputs. Finally, if activation is not None, it is applied to the outputs as well. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 128, 3) for 128x128 RGB pictures in data_format=""channels_last"". You can use None when a dimension has variable size.","2D convolution layer (e.g. spatial convolution over images). This layer creates a convolution kernel that is convolved with the layer input to produce a tensor of outputs. If use_bias is True, a bias vector is created and added to the outputs. Finally, if activation is not None, it is applied to the outputs as well. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 128, 3) for 128x128 RGB pictures in data_format=""channels_last"". You can use None when a dimension has variable size.",A layer that implements 2D convolution (e.g. spatial convolution over images).,A convolutional layer that implements 2D convolution (e.g. spatial convolution over images).,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv2D,,,AIO:ConvolutionalLayer +AIO:Convolution3DLayer,AIO:LayerSubset,Conv3D Layer|nn.Conv3D|Conv3D|Convolution3D|Convolution3D,,Convolution3D Layer,FALSE,0.88,"3D convolution layer (e.g. spatial convolution over volumes). This layer creates a convolution kernel that is convolved with the layer input to produce a tensor of outputs. If use_bias is True, a bias vector is created and added to the outputs. Finally, if activation is not None, it is applied to the outputs as well. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 128, 128, 1) for 128x128x128 volumes with a single channel, in data_format=""channels_last"".","3D convolution layer (e.g. spatial convolution over volumes). This layer creates a convolution kernel that is convolved with the layer input to produce a tensor of outputs. If use_bias is True, a bias vector is created and added to the outputs. Finally, if activation is not None, it is applied to the outputs as well. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 128, 128, 1) for 128x128x128 volumes with a single channel, in data_format=""channels_last"".",A layer that implements 3D convolution (e.g. spatial convolution over volumes).,A convolutional layer that implements 3D convolution (e.g. spatial convolution over volumes).,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv3D,,,AIO:ConvolutionalLayer +AIO:ConvLSTM1DLayer,AIO:LayerSubset,,,ConvLSTM1D Layer,FALSE,0.65,"1D Convolutional LSTM. Similar to an LSTM layer, but the input transformations and recurrent transformations are both convolutional.","1D Convolutional LSTM. Similar to an LSTM layer, but the input transformations and recurrent transformations are both convolutional.",A convolutional layer that implements a 1D Convolutional LSTM similar to an LSTM but with convolutional input and recurrent transformations.,A convolutional layer that implements a 1D Convolutional LSTM similar to an LSTM but with convolutional input and recurrent transformations.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ConvLSTM1D,,,AIO:ConvolutionalLayer +AIO:ConvLSTM2DLayer,AIO:LayerSubset,,,ConvLSTM2D Layer,FALSE,0.65,"2D Convolutional LSTM. Similar to an LSTM layer, but the input transformations and recurrent transformations are both convolutional.","2D Convolutional LSTM. Similar to an LSTM layer, but the input transformations and recurrent transformations are both convolutional.",A convolutional layer that implements a 2D Convolutional LSTM similar to an LSTM but with convolutional input and recurrent transformations.,A convolutional layer that implements a 2D Convolutional LSTM similar to an LSTM but with convolutional input and recurrent transformations.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ConvLSTM2D,,,AIO:ConvolutionalLayer +AIO:ConvLSTM3DLayer,AIO:LayerSubset,,,ConvLSTM3D Layer,FALSE,0.65,"3D Convolutional LSTM. Similar to an LSTM layer, but the input transformations and recurrent transformations are both convolutional.","3D Convolutional LSTM. Similar to an LSTM layer, but the input transformations and recurrent transformations are both convolutional.",A convolutional layer that implements a 3D Convolutional LSTM similar to an LSTM but with convolutional input and recurrent transformations.,A convolutional layer that implements a 3D Convolutional LSTM similar to an LSTM but with convolutional input and recurrent transformations.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ConvLSTM3D,,,AIO:ConvolutionalLayer +AIO:Convolution1DTransposeLayer,AIO:LayerSubset,Conv1DTranspose Layer|nn.ConvTranspose1D|ConvTranspose1D|Convolution1DTranspose|Convolution1DTranspose,,Convolution1DTranspose Layer,FALSE,0.94,"Transposed convolution layer (sometimes called Deconvolution). The need for transposed convolutions generally arises from the desire to use a transformation going in the opposite direction of a normal convolution, i.e., from something that has the shape of the output of some convolution to something that has the shape of its input while maintaining a connectivity pattern that is compatible with said convolution. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 3) for data with 128 time steps and 3 channels.","Transposed convolution layer (sometimes called Deconvolution). The need for transposed convolutions generally arises from the desire to use a transformation going in the opposite direction of a normal convolution, i.e., from something that has the shape of the output of some convolution to something that has the shape of its input while maintaining a connectivity pattern that is compatible with said convolution. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 3) for data with 128 time steps and 3 channels.",A layer that implements transposed 1D convolution sometimes called deconvolution.,A convolutional layer that implements transposed 1D convolution sometimes called deconvolution.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv1DTranspose,,,AIO:ConvolutionalLayer +AIO:Convolution2DTransposeLayer,AIO:LayerSubset,Conv2DTranspose Layer|nn.ConvTranspose2D|ConvTranspose2D|Convolution2DTranspose|Convolution2DTranspose,,Convolution2DTranspose Layer,FALSE,0.82,Transposed convolution layer (sometimes called Deconvolution).,Transposed convolution layer (sometimes called Deconvolution).,A layer that implements transposed 2D convolution,A convolutional layer that implements transposed 2D convolution,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv2DTranspose,,,AIO:ConvolutionalLayer +AIO:Convolution3DTransposeLayer,AIO:LayerSubset,Conv3DTranspose Layer|nn.ConvTranspose3D|ConvTranspose3D|Convolution3DTranspose|Convolution3DTranspose,,Convolution3DTranspose Layer,FALSE,0.95,"Transposed convolution layer (sometimes called Deconvolution). The need for transposed convolutions generally arises from the desire to use a transformation going in the opposite direction of a normal convolution, i.e., from something that has the shape of the output of some convolution to something that has the shape of its input while maintaining a connectivity pattern that is compatible with said convolution. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 128, 128, 3) for a 128x128x128 volume with 3 channels if data_format=""channels_last"".","Transposed convolution layer (sometimes called Deconvolution). The need for transposed convolutions generally arises from the desire to use a transformation going in the opposite direction of a normal convolution, i.e., from something that has the shape of the output of some convolution to something that has the shape of its input while maintaining a connectivity pattern that is compatible with said convolution. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 128, 128, 3) for a 128x128x128 volume with 3 channels if data_format=""channels_last"".",A layer that implements transposed 3D convolution,A convolutional layer that implements transposed 3D convolution,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv3DTranspose,,,AIO:ConvolutionalLayer +AIO:DepthwiseConv1DLayer,AIO:LayerSubset,,,DepthwiseConv1D Layer,FALSE,0.95,"Depthwise 1D convolution. Depthwise convolution is a type of convolution in which each input channel is convolved with a different kernel (called a depthwise kernel). You can understand depthwise convolution as the first step in a depthwise separable convolution. It is implemented via the following steps: Split the input into individual channels. Convolve each channel with an individual depthwise kernel with depth_multiplier output channels. Concatenate the convolved outputs along the channels axis. Unlike a regular 1D convolution, depthwise convolution does not mix information across different input channels. The depth_multiplier argument determines how many filter are applied to one input channel. As such, it controls the amount of output channels that are generated per input channel in the depthwise step.","Depthwise 1D convolution. Depthwise convolution is a type of convolution in which each input channel is convolved with a different kernel (called a depthwise kernel). You can understand depthwise convolution as the first step in a depthwise separable convolution. It is implemented via the following steps: Split the input into individual channels. Convolve each channel with an individual depthwise kernel with depth_multiplier output channels. Concatenate the convolved outputs along the channels axis. Unlike a regular 1D convolution, depthwise convolution does not mix information across different input channels. The depth_multiplier argument determines how many filter are applied to one input channel. As such, it controls the amount of output channels that are generated per input channel in the depthwise step.",A layer that performs depthwise 1D convolution,A convolutional layer that performs depthwise 1D convolution,https://www.tensorflow.org/api_docs/python/tf/keras/layers/DepthwiseConv1D,,,AIO:ConvolutionalLayer +AIO:DepthwiseConv2DLayer,AIO:LayerSubset,,,DepthwiseConv2D Layer,FALSE,0.89,Depthwise 2D convolution.,Depthwise 2D convolution.,A layer that performs depthwise 2D convolution,A convolutional layer that performs depthwise 2D convolution,https://www.tensorflow.org/api_docs/python/tf/keras/layers/DepthwiseConv2D,,,AIO:ConvolutionalLayer +AIO:SeparableConvolution1DLayer,AIO:LayerSubset,SeparableConv1D Layer,,SeparableConvolution1D Layer,FALSE,0.84,"Depthwise separable 1D convolution. This layer performs a depthwise convolution that acts separately on channels, followed by a pointwise convolution that mixes channels. If use_bias is True and a bias initializer is provided, it adds a bias vector to the output. It then optionally applies an activation function to produce the final output.a","Depthwise separable 1D convolution. This layer performs a depthwise convolution that acts separately on channels, followed by a pointwise convolution that mixes channels. If use_bias is True and a bias initializer is provided, it adds a bias vector to the output. It then optionally applies an activation function to produce the final output.a",A layer that performs depthwise separable 1D convolution.,A convolutional layer that performs depthwise separable 1D convolution.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/SeparableConv1D,,,AIO:ConvolutionalLayer +AIO:SeparableConvolution2DLayer,AIO:LayerSubset,SeparableConv2D Layer,,SeparableConvolution2D Layer,FALSE,0.94,"Depthwise separable 2D convolution. Separable convolutions consist of first performing a depthwise spatial convolution (which acts on each input channel separately) followed by a pointwise convolution which mixes the resulting output channels. The depth_multiplier argument controls how many output channels are generated per input channel in the depthwise step. Intuitively, separable convolutions can be understood as a way to factorize a convolution kernel into two smaller kernels, or as an extreme version of an Inception block.","Depthwise separable 2D convolution. Separable convolutions consist of first performing a depthwise spatial convolution (which acts on each input channel separately) followed by a pointwise convolution which mixes the resulting output channels. The depth_multiplier argument controls how many output channels are generated per input channel in the depthwise step. Intuitively, separable convolutions can be understood as a way to factorize a convolution kernel into two smaller kernels, or as an extreme version of an Inception block.",A layer that performs depthwise separable 2D convolution.,A convolutional layer that performs depthwise separable 2D convolution.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/SeparableConv2D,,,AIO:ConvolutionalLayer +AIO:FeatureExtraction,AIO:PreprocessingSubset,Attribute Extraction|Feature Isolation,Syntactic information|Semantic embeddings,Feature Extraction,FALSE,0.07,"The process of transforming raw data into a set of measurable characteristics that can be used as input for machine learning algorithms, enhancing the ability to make accurate predictions.",,The process of transforming raw data into a set of measurable characteristics that can be used as input for machine learning algorithms enhancing the ability to make accurate predictions.,"A data enhancement that transforms raw data into a set of measurable characteristics that can be used as input for machine learning algorithms, enhancing the ability to make accurate predictions.",,,,AIO:DataEnhancement +AIO:DataAugmentation,AIO:PreprocessingSubset,Data Expansion|Data Enrichment,Paraphrasing|Synonym replacement,Data Augmentation,FALSE,0.24,"A technique used to increase the diversity and quantity of training data by applying various transformations such as rotation, scaling, flipping, and cropping to existing data samples, enhancing the robustness and performance of machine learning models.",,A technique used to increase the diversity and quantity of training data by applying various transformations such as rotation scaling flipping and cropping to existing data samples enhancing the robustness and performance of machine learning models.,A data enhancement used to increase the diversity and quantity of training data by applying various transformations such as rotation scaling flipping and cropping to existing data samples enhancing the robustness and performance of machine learning models.,,,,AIO:DataEnhancement +AIO:Tokenization,AIO:PreprocessingSubset,Lexical Analysis|Text Segmentation,,Tokenization,FALSE,0.21,"The process of converting a sequence of text into smaller, meaningful units called tokens, typically words or subwords, for the purpose of analysis or processing by language models.",,The process of converting a sequence of text into smaller meaningful units called tokens typically words or subwords for the purpose of analysis or processing by language models.,A data preparation that converts a sequence of text into smaller meaningful units called tokens typically words or subwords for the purpose of analysis or processing by language models.,,,,AIO:DataPreparation +AIO:SubwordSegmentation,AIO:PreprocessingSubset,Fragmentation|Part-word Division,Byte Pair Encoding|SentencePiece,Subword Segmentation,FALSE,0.12,"The process of dividing text into subword units, which are smaller than words but larger than individual characters, to improve the efficiency and effectiveness of natural language processing models by capturing meaningful subunits of words.",,The process of dividing text into subword units which are smaller than words but larger than individual characters to improve the efficiency and effectiveness of natural language processing models by capturing meaningful subunits of words.,A data preparation that divides text into subword units which are smaller than words but larger than individual characters to improve the efficiency and effectiveness of natural language processing models by capturing meaningful subunits of words.,,,,AIO:DataPreparation +AIO:VocabularyReduction,AIO:PreprocessingSubset,Vocabulary Condensation|Lexical Simplification|Lexicon Pruning,,Vocabulary Reduction,FALSE,0.11,"The technique of limiting the number of unique tokens in a language model’s vocabulary by merging or eliminating less frequent tokens, thereby optimizing computational efficiency and resource usage.",,The technique of limiting the number of unique tokens in a language model's vocabulary by merging or eliminating less frequent tokens thereby optimizing computational efficiency and resource usage.,A data preparation that limits the number of unique tokens in a language model's vocabulary by merging or eliminating less frequent tokens thereby optimizing computational efficiency and resource usage.,,,,AIO:DataPreparation +AIO:Cleaning,AIO:PreprocessingSubset,Data Cleansing|Standardization,Data cleaning|Text normalization,Cleaning,FALSE,0.17,"The process of removing noise, inconsistencies, and irrelevant information from data to enhance its quality and prepare it for analysis or further processing.",,The process of removing noise inconsistencies and irrelevant information from data to enhance its quality and prepare it for analysis or further processing.,A data preparation that removes noise inconsistencies and irrelevant information from data to enhance its quality and prepare it for analysis or further processing.,,,,AIO:DataPreparation +AIO:Normalization,AIO:PreprocessingSubset,,,Normalization,FALSE,#N/A,"The technique of transforming data into a standard format or scale, typically to reduce redundancy and improve consistency, often involving the adjustment of values measured on different scales to a common scale.",,The technique of transforming data into a standard format or scale typically to reduce redundancy and improve consistency often involving the adjustment of values measured on different scales to a common scale.,A data preparation that transforms data into a standard format or scale typically to reduce redundancy and improve consistency often involving the adjustment of values measured on different scales to a common scale.,,,,AIO:DataPreparation +AIO:NeuralTuringMachineNetwork,AIO:NetworkSubset,NTM,,Neural Turing Machine Network,FALSE,0.66,"A neural Turing machine (NTM) combines neural network pattern matching with the algorithmic power of programmable computers, using attention mechanisms to interact with external memory for tasks like copying, sorting, and associative recall.",,A deep feedforward network that combines neural network pattern matching with the algorithmic power of programmable computers.,A deep feedforward network that combines neural network pattern matching with the algorithmic power of programmable computers.,https://en.wikipedia.org/wiki/Neural_Turing_machine,"Layers: Input, Hidden, Spiking Hidden, Output",AIO:InputLayer|AIO:HiddenLayer|AIO:SpikingHiddenLayer|AIO:OutputLayer,AIO:DeepFeedForwardNetwork +AIO:RadialBasisNetwork,AIO:NetworkSubset,RBN|Radial Basis Function Network|RBFN,,Radial Basis Network,FALSE,0.55,"Radial basis function networks use radial basis functions as activation functions, effective for pattern recognition and interpolation.",,A deep feedforward network that uses radial basis functions as activation functions for pattern recognition and interpolation.,A deep feedforward network that uses radial basis functions as activation functions for pattern recognition and interpolation.,https://en.wikipedia.org/wiki/Radial_basis_function_network,"Layers: Input, Hidden, Output",AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:DeepFeedForwardNetwork +AIO:ContrastiveLearning,AIO:NetworkSubset,,,Contrastive Learning,FALSE,0.82,"Contrastive learning is a self-supervised learning approach in which the model learns to distinguish between similar and dissimilar pairs of data samples. By maximizing the similarity between positive pairs (similar samples) and minimizing the similarity between negative pairs (dissimilar samples), the model learns to capture meaningful representations of the data. This method is particularly effective for representation learning and is widely used in tasks such as image classification, clustering, and retrieval. Contrastive learning techniques often employ loss functions such as the contrastive loss or the triplet loss to achieve these objectives.","Contrastive learning is a self-supervised learning approach in which the model learns to distinguish between similar and dissimilar pairs of data samples. By maximizing the similarity between positive pairs (similar samples) and minimizing the similarity between negative pairs (dissimilar samples), the model learns to capture meaningful representations of the data. This method is particularly effective for representation learning and is widely used in tasks such as image classification, clustering, and retrieval. Contrastive learning techniques often employ loss functions such as the contrastive loss or the triplet loss to achieve these objectives.",A deep neural network self-supervised learning approach that learns to distinguish between similar and dissimilar data samples.,A deep neural network self-supervised learning approach that learns to distinguish between similar and dissimilar data samples.,https://arxiv.org/abs/2202.14037,,,AIO:DeepNeuralNetwork +AIO:DeepConvolutionalNetwork,AIO:NetworkSubset,DCN|Convolutional Neural Network|CNN|ConvNet,,Deep Convolutional Network,FALSE,0.58,"A deep convolutional network (CNN) is an artificial neural network used to analyze visual imagery, utilizing shared-weight architecture and translation-equivariant feature maps.",,A deep neural network specialized for analyzing visual imagery using shared-weight architecture and translation-equivariant feature maps.,A deep neural network specialized for analyzing visual imagery using shared-weight architecture and translation-equivariant feature maps.,https://en.wikipedia.org/wiki/Convolutional_neural_network,"Layers: Input, Kernel, Convolutional/Pool, Hidden, Output",AIO:InputLayer|AIO:KernelLayer|AIO:ConvolutionalLayer|AIO:PoolingLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:DeepNeuralNetwork +AIO:OneshotLearning,AIO:NetworkSubset,OSL,,One-shot Learning,FALSE,#N/A,,,A deep neural network that classified objects from one or only a few examples.,A deep neural network that classified objects from one or only a few examples.,https://en.wikipedia.org/wiki/One-shot_learning,,,AIO:DeepNeuralNetwork +AIO:DeepActiveLearning,AIO:NetworkSubset,DeepAL,,Deep Active Learning,FALSE,0.33,Combining deep learning and active learning to maximize model performance gain while annotating the fewest samples possible.,,A deep neural network that combines deep learning and active learning to maximize model performance while annotating the fewest samples possible.,A deep neural network that combines deep learning and active learning to maximize model performance while annotating the fewest samples possible.,https://arxiv.org/pdf/2009.00236.pdf,,,AIO:DeepNeuralNetwork +AIO:RepresentationLearning,AIO:NetworkSubset,Feature Learning,,Representation Learning,FALSE,0.41,Discovering representations required for feature detection or classification from raw data.,Discovering representations required for feature detection or classification from raw data.,A deep neural network that discovers representations required for feature detection or classification from raw data.,A deep neural network that discovers representations required for feature detection or classification from raw data.,https://en.wikipedia.org/wiki/Feature_learning,,,AIO:DeepNeuralNetwork +AIO:ResidualNeuralNetwork,AIO:NetworkSubset,ResNN|Deep Residual Network|ResNet|DRN,,Residual Neural Network,FALSE,0.35,"A residual neural network (ResNet) employs skip connections to bypass certain layers, facilitating the learning of residual functions.",,A deep neural network that employs skip connections to bypass layers facilitating learning of residual functions.,A deep neural network that employs skip connections to bypass layers facilitating learning of residual functions.,https://en.wikipedia.org/wiki/Residual_neural_network,"Layers: Input, Weight, BN, ReLU, Weight, BN, Addition, ReLU",AIO:InputLayer|AIO:WeightedLayer|AIO:BatchNormalizationLayer|AIO:ReLULayer|AIO:AdditionLayer,AIO:DeepNeuralNetwork +AIO:MetricLearning,AIO:NetworkSubset,Distance Metric Learning,,Metric Learning,FALSE,0.47,Learning a representation function that maps objects into an embedded space.,Learning a representation function that maps objects into an embedded space.,A deep neural network that learns a representation function mapping objects into an embedded space.,A deep neural network that learns a representation function mapping objects into an embedded space.,https://paperswithcode.com/task/metric-learning,,,AIO:DeepNeuralNetwork +AIO:GeneralizedFewshotLearning,AIO:NetworkSubset,GFSL,,Generalized Few-shot Learning,FALSE,#N/A,,,"A deep neural network that learns novel classes from few samples per class, preventing catastrophic forgetting of base classes and ensuring classifier calibration.","A deep neural network that learns novel classes from few samples per class, preventing catastrophic forgetting of base classes and ensuring classifier calibration.",https://paperswithcode.com/paper/generalized-and-incremental-few-shot-learning/review/,,,AIO:DeepNeuralNetwork +AIO:ContinualLearning,AIO:NetworkSubset,Incremental Learning|Life-Long Learning,,Continual Learning,FALSE,0.48,"Learning a model for sequential tasks without forgetting knowledge from preceding tasks, with no access to old task data during new task training.","Learning a model for sequential tasks without forgetting knowledge from preceding tasks, with no access to old task data during new task training.",A deep neural network that learns sequential tasks without forgetting knowledge from preceding tasks and without access to old task data during new task training.,A deep neural network that learns sequential tasks without forgetting knowledge from preceding tasks and without access to old task data during new task training.,https://paperswithcode.com/task/continual-learning,,,AIO:DeepNeuralNetwork +AIO:GraphConvolutionalNetwork,AIO:NetworkSubset,GCN,,Graph Convolutional Network,FALSE,0.68,"A graph convolutional network (GCN) operates directly on graph structures, utilizing their structural information for tasks like node classification and graph clustering.",,A deep neural network that operates directly on graph structures utilizing structural information.,A deep neural network that operates directly on graph structures utilizing structural information.,https://arxiv.org/abs/1609.02907,"Layers: Input, Hidden, Hidden, Output",AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:DeepNeuralNetwork +AIO:ZeroshotLearning,AIO:NetworkSubset,ZSL,,Zero-shot Learning,FALSE,#N/A,,,A deep neural network that predicts classes at test time from classes not observed during training.,A deep neural network that predicts classes at test time from classes not observed during training.,https://en.wikipedia.org/wiki/Zero-shot_learning,,,AIO:DeepNeuralNetwork +AIO:MultimodalDeepLearning,AIO:NetworkSubset,,,Multimodal Deep Learning,FALSE,0.63,Creating models that process and link information using various modalities.,Creating models that process and link information using various modalities.,A deep neural network that processes and links information using various modalities.,A deep neural network that processes and links information using various modalities.,https://arxiv.org/abs/2105.11087,,,AIO:DeepNeuralNetwork +AIO:DeepFeedForwardNetwork,AIO:NetworkSubset,DFF|Multilayer Perceptoron|MLP,,Deep Feed-Forward Network,FALSE,#N/A,,,A deep neural network that processes information in one direction—from input nodes through hidden nodes to output nodes—without cycles or loops.,A deep neural network that processes information in one direction—from input nodes through hidden nodes to output nodes—without cycles or loops.,https://en.wikipedia.org/wiki/Feedforward_neural_network,"Layers: Input, Hidden, Output",AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:DeepNeuralNetwork +AIO:RecursiveNeuralNetwork,AIO:NetworkSubset,RecuNN|RvNN,,Recursive Neural Network,FALSE,0.33,A recursive neural network applies the same set of weights recursively over structured input to generate structured or scalar predictions.,,A deep neural network that recursively applies weights over structured input to generate structured or scalar predictions.,A deep neural network that recursively applies weights over structured input to generate structured or scalar predictions.,https://en.wikipedia.org/wiki/Recursive_neural_network,,,AIO:DeepNeuralNetwork +AIO:DeepTransferLearning,AIO:NetworkSubset,,,Deep Transfer Learning,FALSE,0.29,Relaxing the hypothesis that training data must be independent and identically distributed (i.i.d.) with test data to address insufficient training data.,,A deep neural network that relaxes the hypothesis that training data must be independent and identically distributed with test data to address insufficient training data.,A deep neural network that relaxes the hypothesis that training data must be independent and identically distributed with test data to address insufficient training data.,https://arxiv.org/abs/1808.01974,,,AIO:DeepNeuralNetwork +AIO:DeconvolutionalNetwork,AIO:NetworkSubset,DN,,Deconvolutional Network,FALSE,0.78,Deconvolutional networks allow unsupervised construction of hierarchical image representations for tasks such as denoising and feature extraction for object recognition.,,A deep neural network that uses deconvolution for unsupervised construction of hierarchical image representations.,A deep neural network that uses deconvolution for unsupervised construction of hierarchical image representations.,https://ieeexplore.ieee.org/document/5539957,"Layers: Input, Kernel, Convolutional/Pool, Output",AIO:InputLayer|AIO:KernelLayer|AIO:ConvolutionalLayer|AIO:PoolingLayer|AIO:OutputLayer,AIO:DeepNeuralNetwork +AIO:RecursiveLanguageModel,AIO:ModelSubset,RLM,Compositional generalization,Recursive Language Model,FALSE,0.14,"A recursive language model uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions, improving systematic generalization abilities.",,A language model that uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions improving systematic generalization abilities.,A deep neural network that uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions improving systematic generalization abilities.,https://en.wikipedia.org/wiki/Recurrent_neural_network,"Layers: Input, Memory Cell, Output",AIO:InputLayer|AIO:MemoryCellLayer|AIO:OutputLayer,AIO:DeepNeuralNetwork +AIO:TransformerNetwork,AIO:NetworkSubset,,,Transformer Network,FALSE,0.63,"A transformer network utilizes attention mechanisms to weigh the significance of each part of the input data, widely used in natural language processing (NLP) and computer vision (CV).","A transformer network utilizes attention mechanisms to weigh the significance of each part of the input data, widely used in natural language processing (NLP) and computer vision (CV).",A deep neural network that utilizes attention mechanisms to weigh the significance of input data.,A deep neural network that utilizes attention mechanisms to weigh the significance of input data.,https://en.wikipedia.org/wiki/Transformer_(machine_Learning_model),,,AIO:DeepNeuralNetwork +AIO:FederatedLearning,AIO:NetworkSubset,,,Federated Learning,FALSE,0.41,Training an algorithm across multiple decentralized edge devices or servers holding local data samples without exchanging them.,Training an algorithm across multiple decentralized edge devices or servers holding local data samples without exchanging them.,A deep neural network trained across decentralized edge devices or servers holding local data samples without exchanging them.,A deep neural network trained across decentralized edge devices or servers holding local data samples without exchanging them.,https://en.wikipedia.org/wiki/Federated_learning,,,AIO:DeepNeuralNetwork +AIO:IncremenetalFewshotLearning,AIO:NetworkSubset,IFSL,,Incremenetal Few-shot Learning,FALSE,#N/A,,,"A deep neural network trained on a base set of classes and then presented with novel classes, each with few labeled examples.","A deep neural network trained on a base set of classes and then presented with novel classes, each with few labeled examples.",https://arxiv.org/abs/1810.07218,,,AIO:DeepNeuralNetwork +AIO:tDistributedStochasticNeighborembedding,AIO:MachineLearningSubset,tSNE|t-SNE,,t-Distributed Stochastic Neighbor embedding,FALSE,#N/A,,,A dimensionality reduction for visualizing high-dimensional data by giving each datapoint a location in a two or three-dimensional map.,A dimensionality reduction for visualizing high-dimensional data by giving each datapoint a location in a two or three-dimensional map.,https://en.wikipedia.org/wiki/T-distributed_stochastic_neighbor_embedding,,,AIO:DimensionalityReduction +AIO:ManifoldLearning,AIO:MachineLearningSubset,,,Manifold Learning,FALSE,0.25,Methods based on the assumption that observed data lie on a low-dimensional manifold embedded in a higher-dimensional space.,,A dimensionality reduction method based on the assumption that observed data lie on a low-dimensional manifold embedded in a higher-dimensional space.,A dimensionality reduction method based on the assumption that observed data lie on a low-dimensional manifold embedded in a higher-dimensional space.,https://arxiv.org/abs/2011.01307,,,AIO:DimensionalityReduction +AIO:PrincipalComponentAnalysis,AIO:MachineLearningSubset,PCA,,Principal Component Analysis,FALSE,0.32,"A method for analyzing large datasets with high-dimensional features per observation, increasing data interpretability while preserving maximum information and enabling visualization of multidimensional data.",,A dimensionality reduction method for analyzing large datasets with high-dimensional features per observation increasing data interpretability while preserving maximum information and enabling visualization.,A dimensionality reduction method for analyzing large datasets with high-dimensional features per observation increasing data interpretability while preserving maximum information and enabling visualization.,https://en.wikipedia.org/wiki/Principal_component_analysis,,,AIO:DimensionalityReduction +AIO:MultidimensionalScaling,AIO:MachineLearningSubset,MDS,,Multidimensional Scaling,FALSE,0.08,A method that translates information about the pairwise distances among a set of objects or individuals into a configuration of points mapped into an abstract Cartesian space.,,A dimensionality reduction method that translates information about the pairwise distances among a set of objects or individuals into a configuration of points mapped into an abstract Cartesian space.,A dimensionality reduction method that translates information about the pairwise distances among a set of objects or individuals into a configuration of points mapped into an abstract Cartesian space.,https://en.wikipedia.org/wiki/Multidimensional_scaling,,,AIO:DimensionalityReduction +AIO:ExtremeLearningMachine,AIO:NetworkSubset,ELM,,Extreme Learning Machine,FALSE,0.76,"Extreme learning machines are feedforward neural networks with randomly assigned hidden node parameters that are not updated, learning output weights in a single step.",,A feedback network with randomly assigned hidden nodes that are not updated during training.,A feedback network with randomly assigned hidden nodes that are not updated during training.,https://en.wikipedia.org/wiki/Extreme_Learning_machine,"Layers: Input, Hidden, Output",AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:FeedbackNetwork +AIO:MachineLearningTask,AIO:MachineLearningSubset,,Machine Learning,Machine Learning Task,FALSE,#N/A,A field of inquiry devoted to understanding and building methods that learn from data to improve performance on a set of tasks.,,#N/A,A field of inquiry devoted to understanding and building methods that learn from data to improve performance on a set of tasks.,https://doi.org/10.6028/NIST.SP.1270,,,owl:Thing +AIO:GraphConvolutionalPolicyNetwork,AIO:NetworkSubset,GPCN,,Graph Convolutional Policy Network,FALSE,0.39,"A graph convolutional policy network (GCPN) generates goal-directed graphs using a graph convolutional network and reinforcement learning, optimizing for domain-specific rewards and adversarial loss",,A graph convolutional network that generates goal-directed graphs using reinforcement learning and optimizing for rewards and adversarial loss.,A graph convolutional network that generates goal-directed graphs using reinforcement learning and optimizing for rewards and adversarial loss.,https://arxiv.org/abs/1806.02473,"Layers: Input, Hidden, Hidden, Policy, Output",AIO:InputLayer|AIO:HiddenLayer|AIO:PolicyLayer|AIO:OutputLayer,AIO:GraphConvolutionalNetwork +AIO:FundingBias,AIO:BiasSubset,,,Funding Bias,FALSE,0.14,Bias arising when biased results are reported to support or satisfy the funding agency or financial supporter of a research study.,,A bias arising when biased results are reported to support or satisfy the funding agency or financial supporter of a research study.,A group bias arising when biased results are reported to support or satisfy the funding agency or financial supporter of a research study.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:GroupBias +AIO:DeploymentBias,AIO:BiasSubset,,,Deployment Bias,FALSE,0.36,"Arises when systems are used as decision aids for humans, since the human intermediary may act on predictions in ways that are typically not modeled in the system. However, it is still individuals using the deployed system.",,A bias arising when systems are used as decision aids for humans since the human intermediary may act on predictions in ways that are typically not modeled in the system.,A group bias arising when systems are used as decision aids for humans since the human intermediary may act on predictions in ways that are typically not modeled in the system.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:GroupBias +AIO:GroupthinkBias,AIO:BiasSubset,Groupthink,,Groupthink Bias,FALSE,0.00,A psychological phenomenon where people in a group make non-optimal decisions due to a desire to conform or fear of dissent.,,A psychological phenomenon where people in a group make non-optimal decisions due to a desire to conform or fear of dissent.,A group bias in which people in a group make non-optimal decisions due to a desire to conform or fear of dissent.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:GroupBias +AIO:ProbabilisticHiddenLayer,AIO:LayerSubset,,,Probabilistic Hidden Layer,FALSE,#N/A,,,A hidden layer that estimates the probability of a sample being within a certain category.,A hidden layer that estimates the probability of a sample being within a certain category.,,,,AIO:HiddenLayer +AIO:SpikingHiddenLayer,AIO:LayerSubset,,,Spiking Hidden Layer,FALSE,#N/A,,,"A hidden layer that makes connections to an additional, heterogeneous hidden layer; modeled after biological neural networks.","A hidden layer that makes connections to an additional, heterogeneous hidden layer; modeled after biological neural networks.",https://doi.org/10.1016/S0893-6080(97)00011-7,,,AIO:HiddenLayer +AIO:GroupBias,AIO:BiasSubset,In-group Favoritism|In-group–out-group Bias|In-group bias|Intergroup bias|In-group preference,,Group Bias,FALSE,0.48,"Favoring members of one's in-group over out-group members, expressed in evaluation, resource allocation, and other ways.","Favoring members of one's in-group over out-group members, expressed in evaluation, resource allocation, and other ways.",A bias characterized by favoring members of one's in-group over out-group members expressed in evaluation resource allocation and other ways.,A systemic bias characterized by favoring members of one's in-group over out-group members expressed in evaluation resource allocation and other ways.,https://en.wikipedia.org/wiki/In-group_favoritism,,,AIO:SystemicBias +AIO:LargeLanguageModel,AIO:ModelSubset,LLM,,Large Language Model,FALSE,0.07,"A language model consisting of a neural network with many parameters (typically billions of weights or more), trained on large quantities of unlabeled text using self-supervised learning or semi-supervised learning.",,A language model consisting of a neural network with many parameters (typically billions of weights or more) trained on large quantities of unlabeled text using self-supervised learning or semi-supervised learning.,A language model consisting of a neural network with many parameters (typically billions of weights or more) trained on large quantities of unlabeled text using self-supervised learning or semi-supervised learning.,https://en.wikipedia.org/wiki/Large_language_model,,,AIO:LanguageModel +AIO:ModularLanguageModel,AIO:ModelSubset,Modular LM,,Modular Language Model,FALSE,0.07,"A modular language model consists of multiple specialized components or skills that can be dynamically composed and recombined to solve complex tasks, mimicking the modular structure of human cognition.",,A language model that consists of multiple specialized components or skills that can be dynamically composed and recombined to solve complex tasks mimicking the modular structure of human cognition.,A language model that consists of multiple specialized components or skills that can be dynamically composed and recombined to solve complex tasks mimicking the modular structure of human cognition.,https://arxiv.org/abs/2302.11529v2,,,AIO:LanguageModel +AIO:GenerativeLanguageInterface,AIO:ModelSubset,,Interactive generation,Generative Language Interface,FALSE,0.23,"A generative language interface enables users to engage in an interactive dialogue with an LLM, providing feedback to guide and refine the generated outputs iteratively.",,A language model that enables users to engage in an interactive dialogue with an LLM providing feedback to guide and refine the generated outputs iteratively.,A language model that enables users to engage in an interactive dialogue with an LLM providing feedback to guide and refine the generated outputs iteratively.,,,,AIO:LanguageModel +AIO:AutoregressiveLanguageModel,AIO:ModelSubset,,generative language model|sequence-to-sequence model,Autoregressive Language Model,FALSE,0.28,"An autoregressive language model generates text sequentially, predicting one token at a time based on the previously generated tokens. It excels at natural language generation tasks by modeling the probability distribution over sequences of tokens.",,A language model that generates text sequentially predicting one token at a time based on the previously generated tokens excelling at natural language generation tasks by modeling the probability distribution over sequences of tokens.,A language model that generates text sequentially predicting one token at a time based on the previously generated tokens excelling at natural language generation tasks by modeling the probability distribution over sequences of tokens.,,,,AIO:LanguageModel +AIO:MaskedLanguageModel,AIO:ModelSubset,,bidirectional encoder|denoising autoencoder,Masked Language Model,FALSE,0.16,"A masked language model is trained to predict randomly masked tokens in a sequence, based on the remaining unmasked tokens. This allows it to build deep bidirectional representations that can be effectively transferred to various NLP tasks via fine-tuning.",,A language model that is trained to predict randomly masked tokens in a sequence based on the remaining unmasked tokens allowing it to build deep bidirectional representations that can be effectively transferred to various NLP tasks via fine-tuning.,A language model that is trained to predict randomly masked tokens in a sequence based on the remaining unmasked tokens allowing it to build deep bidirectional representations that can be effectively transferred to various NLP tasks via fine-tuning.,,,,AIO:LanguageModel +AIO:MultimodalLanguageModel,AIO:ModelSubset,Mulimodal LM,,Multimodal Language Model,FALSE,0.22,"A multimodal language model learns joint representations across different modalities like text, vision, and audio in an end-to-end fashion for better cross-modal understanding and generation.",,A language model that learns joint representations across different modalities like text vision and audio in an end-to-end fashion for better cross-modal understanding and generation.,A language model that learns joint representations across different modalities like text vision and audio in an end-to-end fashion for better cross-modal understanding and generation.,https://arxiv.org/abs/2205.12630,,,AIO:LanguageModel +AIO:GraphLanguageModel,AIO:ModelSubset,Graph LM,Structured representations,Graph Language Model,FALSE,0.17,"A graph language model operates over structured inputs or outputs represented as graphs, enabling reasoning over explicit relational knowledge representations during language tasks.",,A language model that operates over structured inputs or outputs represented as graphs enabling reasoning over explicit relational knowledge representations during language tasks.,A language model that operates over structured inputs or outputs represented as graphs enabling reasoning over explicit relational knowledge representations during language tasks.,https://arxiv.org/abs/2401.07105,,,AIO:LanguageModel +AIO:HierarchicalLanguageModel,AIO:ModelSubset,Hierarchical LM,multi-scale representations,Hierarchical Language Model,FALSE,0.10,"A hierarchical language model represents language at multiple levels of granularity, learning hierarchical representations that capture both low-level patterns and high-level abstractions.",,A language model that represents language at multiple levels of granularity learning hierarchical representations that capture both low-level patterns and high-level abstractions.,A language model that represents language at multiple levels of granularity learning hierarchical representations that capture both low-level patterns and high-level abstractions.,https://doi.org/10.1016/j.ipm.2024.103698,,,AIO:LanguageModel +AIO:ImplicitLanguageModel,AIO:ModelSubset,Implicit LM,Energy-based models|Token-level scoring,Implicit Language Model,FALSE,0.22,"An implicit language model uses an energy function to score entire sequences instead of factorizing probabilities autoregressively, better capturing global properties and long-range dependencies.",,A language model that uses an energy function to score entire sequences instead of factorizing probabilities autoregressively better capturing global properties and long-range dependencies.,A language model that uses an energy function to score entire sequences instead of factorizing probabilities autoregressively better capturing global properties and long-range dependencies.,https://arxiv.org/pdf/2303.16189,,,AIO:LanguageModel +AIO:RecursiveLLM,AIO:ModelSubset,Recursive Large Language Model|Self-Attending Large Language Model,self-attention|iterative refinement,Recursive LLM,FALSE,0.18,"A recursive language model uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions, improving systematic generalization abilities.",,A large language model that uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions improving systematic generalization abilities.,A language model that uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions improving systematic generalization abilities.,https://doi.org/10.1609/aaai.v33i01.33017450,,,AIO:LanguageModel +AIO:TransformerLanguageModel,AIO:ModelSubset,Transformer LM,,Transformer Language Model,FALSE,0.19,"A transformer LM is a neural network model that uses the transformer architecture based on multi-head attention mechanisms, allowing it to contextualize tokens within a context window for effective language understanding and generation.",,A language model that uses the transformer architecture based on multi-head attention mechanisms allowing it to contextualize tokens within a context window for effective language understanding and generation.,A language model that uses the transformer architecture based on multi-head attention mechanisms allowing it to contextualize tokens within a context window for effective language understanding and generation.,https://arxiv.org/abs/1706.03762,,,AIO:LanguageModel +AIO:FactoredLanguageModel,AIO:ModelSubset,Factorized Language Model,,Factored Language Model,FALSE,0.38,"A factored language model views each word as a vector of multiple factors, such as part-of-speech, morphology, and semantics, to improve language modeling.",,A language model that views each word as a vector of multiple factors such as part-of-speech morphology and semantics to improve language modeling.,A language model that views each word as a vector of multiple factors such as part-of-speech morphology and semantics to improve language modeling.,https://en.wikipedia.org/wiki/Factored_language_model,,,AIO:LanguageModel +AIO:EncoderDecoderLLM,AIO:ModelSubset,,,Encoder-Decoder LLM,FALSE,#N/A,,,"The LLM introduced in the ""Attention Is All You Need"" paper. The encoder processes the input sequence to generate a hidden representation summarizing the input information, while the decoder uses this hidden representation to generate the desired output sequence.","A large language model introduced in the ""Attention Is All You Need"" paper. The encoder processes the input sequence to generate a hidden representation summarizing the input information, while the decoder uses this hidden representation to generate the desired output sequence.",https://www.practicalai.io/understanding-transformer-model-architectures/#:~:text=Encoder,,,AIO:LargeLanguageModel +AIO:PersonalizedLLM,AIO:ModelSubset,Personalized Large Language Model,user adaptation LLM,Personalized LLM,FALSE,0.38,"A personalized LLM adapts its language modeling and generation to the preferences, style, and persona of individual users or audiences.",,A large language model that adapts its language modeling and generation to the preferences style and persona of individual users or audiences.,A large language model that adapts its language modeling and generation to the preferences style and persona of individual users or audiences.,,,,AIO:LargeLanguageModel +AIO:ControllableLLM,AIO:ModelSubset,Controllable Large Language Model,conditional generation|guided generation,Controllable LLM,FALSE,0.43,"A controllable LLM allows for explicit control over certain attributes of the generated text, such as style, tone, topic, or other desired characteristics, through conditioning or specialized training objectives.","A controllable LLM allows for explicit control over certain attributes of the generated text, such as style, tone, topic, or other desired characteristics, through conditioning or specialized training objectives.",A large language model that allows for explicit control over certain attributes of the generated text such as style tone topic or other desired characteristics through conditioning or specialized training objectives.,A large language model that allows for explicit control over certain attributes of the generated text such as style tone topic or other desired characteristics through conditioning or specialized training objectives.,,,,AIO:LargeLanguageModel +AIO:EvolutionaryLLM,AIO:ModelSubset,Evolutionary Language Model,evolutionary algorithms|genetic programming,Evolutionary LLM,FALSE,0.38,"An evolutionary LLM applies principles of evolutionary computation to optimize its structure and parameters, evolving over time to improve performance.",,A large language model that applies principles of evolutionary computation to optimize its structure and parameters evolving over time to improve performance.,A large language model that applies principles of evolutionary computation to optimize its structure and parameters evolving over time to improve performance.,,,,AIO:LargeLanguageModel +AIO:ContinualLearningLLM,AIO:ModelSubset,CL-Large Language Model|Continual Learning Large Language Model,lifelong learning|catastrophic forgetting,Continual Learning LLM,FALSE,0.31,A continual learning LLM continually acquires new knowledge and skills over time without forgetting previously learned information. This allows the model to adapt and expand its capabilities as new data becomes available.,,A large language model that continually acquires new knowledge and skills over time without forgetting previously learned information allowing the model to adapt and expand its capabilities as new data becomes available.,A large language model that continually acquires new knowledge and skills over time without forgetting previously learned information allowing the model to adapt and expand its capabilities as new data becomes available.,,,,AIO:LargeLanguageModel +AIO:LifelongLearningLLM,AIO:ModelSubset,Continual Learning LLM|Forever Learning,Catastrophic forgetting|Plasticity-Stability balance,Lifelong Learning LLM,FALSE,0.33,"A lifelong learning LLM continually acquires new knowledge over time without forgetting previously learned information, maintaining a balance between plasticity and stability.",,A large language model that continually acquires new knowledge over time without forgetting previously learned information maintaining a balance between plasticity and stability.,A large language model that continually acquires new knowledge over time without forgetting previously learned information maintaining a balance between plasticity and stability.,,,,AIO:LargeLanguageModel +AIO:FactorizedLLM,AIO:ModelSubset,Factorized Large Language Model|Factorized Learning Assisted with Large Language Model,Conditional masking|Product of experts,Factorized LLM,FALSE,0.21,"A factorized LLM decomposes the full language modeling task into multiple sub-components or experts that each focus on a subset of the information, enabling more efficient scaling.",,A large language model that decomposes the full language modeling task into multiple sub-components or experts that each focus on a subset of the information enabling more efficient scaling.,A large language model that decomposes the full language modeling task into multiple sub-components or experts that each focus on a subset of the information enabling more efficient scaling.,https://doi.org/10.48550/arXiv.2403.12556,,,AIO:LargeLanguageModel +AIO:DatatoTextLLM,AIO:ModelSubset,Meaning representation,,Data-to-Text LLM,FALSE,#N/A,,,"A LLM that generates natural language descriptions from structured data sources like tables, graphs, and knowledge bases, requiring grounding in meaning representations.","A large language model that generates natural language descriptions from structured data sources like tables, graphs, and knowledge bases, requiring grounding in meaning representations.",,,,AIO:LargeLanguageModel +AIO:DifferentiableLLM,AIO:ModelSubset,Differentiable Large Language Model,end-to-end training|fully backpropagable,Differentiable LLM,FALSE,0.31,"A differentiable LLM has an architecture amenable to full end-to-end training via backpropagation, without relying on teacher forcing or unlikelihood training objectives.",,A large language model that has an architecture amenable to full end-to-end training via backpropagation without relying on teacher forcing or unlikelihood training objectives.,A large language model that has an architecture amenable to full end-to-end training via backpropagation without relying on teacher forcing or unlikelihood training objectives.,,,,AIO:LargeLanguageModel +AIO:ReasoningLLM,AIO:ModelSubset,Reasoning Large Language Model|Rational Large Language Model,reasoning|logical inferences,Reasoning LLM,FALSE,0.38,"A reasoning LLM incorporates explicit reasoning capabilities, leveraging logical rules, axioms, or external knowledge to make deductive inferences during language tasks.",,A large language model that incorporates explicit reasoning capabilities leveraging logical rules axioms or external knowledge to make deductive inferences during language tasks.,A large language model that incorporates explicit reasoning capabilities leveraging logical rules axioms or external knowledge to make deductive inferences during language tasks.,https://doi.org/10.18653/v1/2023.acl-long.347,,,AIO:LargeLanguageModel +AIO:EmbodiedLLM,AIO:ModelSubset,Embodied Large Language Model,multimodal grounding,Embodied LLM,FALSE,0.42,"An embodied LLM integrates language with other modalities like vision, audio, and robotics to enable grounded language understanding in real-world environments.","An embodied LLM integrates language with other modalities like vision, audio, and robotics to enable grounded language understanding in real-world environments.",A large language model that integrates language with other modalities like vision audio and robotics to enable grounded language understanding in real-world environments.,A large language model that integrates language with other modalities like vision audio and robotics to enable grounded language understanding in real-world environments.,,,,AIO:LargeLanguageModel +AIO:ExplainableLLM,AIO:ModelSubset,Explainable Language Model|XAI LLM,interpretability|model understanding,Explainable LLM,FALSE,0.39,"An explainable LLM is designed to provide insights into its decision-making process, making it easier for users to understand and trust the model's outputs. It incorporates mechanisms for interpreting and explaining its predictions in human-understandable terms.",,A large language model that is designed to provide insights into its decision-making process making it easier for users to understand and trust the model's outputs by incorporating mechanisms for interpreting and explaining its predictions in human-understandable terms.,A large language model that is designed to provide insights into its decision-making process making it easier for users to understand and trust the model's outputs by incorporating mechanisms for interpreting and explaining its predictions in human-understandable terms.,,,,AIO:LargeLanguageModel +AIO:ReinforcementLearningLLM,AIO:ModelSubset,RL-Large Language Model|Reinforcement Learning Large Language Model,reward modeling|decision transformers,Reinforcement Learning LLM,FALSE,0.43,"An RL-LLM is a language model fine-tuned using reinforcement learning, where the model receives rewards for generating text that satisfies certain desired properties or objectives. This can improve the quality, safety, or alignment of generated text.","An RL-LLM is a language model fine-tuned using reinforcement learning, where the model receives rewards for generating text that satisfies certain desired properties or objectives. This can improve the quality, safety, or alignment of generated text.",A large language model that is fine-tuned using reinforcement learning where the model receives rewards for generating text that satisfies certain desired properties or objectives improving the quality safety or alignment of generated text.,A large language model that is fine-tuned using reinforcement learning where the model receives rewards for generating text that satisfies certain desired properties or objectives improving the quality safety or alignment of generated text.,,,,AIO:LargeLanguageModel +AIO:DialogueLLM,AIO:ModelSubset,Dialogue Large Language Model,conversational AI|multi-turn dialogue,Dialogue LLM,FALSE,0.39,"A dialogue LLM is optimized for engaging in multi-turn conversations, understanding context, and generating relevant, coherent responses continuously over many dialogue turns.",,A large language model that is optimized for engaging in multi-turn conversations understanding context and generating relevant coherent responses continuously over many dialogue turns.,A large language model that is optimized for engaging in multi-turn conversations understanding context and generating relevant coherent responses continuously over many dialogue turns.,,,,AIO:LargeLanguageModel +AIO:CurriculumLearningLLM,AIO:ModelSubset,,Learning progression,Curriculum Learning LLM,FALSE,0.30,"A curriculum learning LLM is trained by presenting learning examples in a meaningful order from simple to complex, mimicking the learning trajectory followed by humans.",,A large language model that is trained by presenting learning examples in a meaningful order from simple to complex mimicking the learning trajectory followed by humans.,A large language model that is trained by presenting learning examples in a meaningful order from simple to complex mimicking the learning trajectory followed by humans.,,,,AIO:LargeLanguageModel +AIO:FederatedLLM,AIO:ModelSubset,Federated Large Language Model,privacy-preserving|decentralized training,Federated LLM,FALSE,0.34,"A federated LLM is trained in a decentralized manner across multiple devices or silos, without directly sharing private data. This enables collaborative training while preserving data privacy and security.",,A large language model that is trained in a decentralized manner across multiple devices or silos without directly sharing private data enabling collaborative training while preserving data privacy and security.,A large language model that is trained in a decentralized manner across multiple devices or silos without directly sharing private data enabling collaborative training while preserving data privacy and security.,,,,AIO:LargeLanguageModel +AIO:MultilingualLLM,AIO:ModelSubset,Multilingual Large Language Model,cross-lingual transfer,Multilingual LLM,FALSE,0.27,"A multilingual LLM is trained on text from multiple languages, learning shared representations that enable zero-shot or few-shot transfer to new languages.",,A large language model that is trained on text from multiple languages learning shared representations that enable zero-shot or few-shot transfer to new languages.,A large language model that is trained on text from multiple languages learning shared representations that enable zero-shot or few-shot transfer to new languages.,,,,AIO:LargeLanguageModel +AIO:UnsupervisedLLM,AIO:ModelSubset,Unsupervised Large Language Model,self-supervised,Unsupervised LLM,FALSE,0.36,"An unsupervised LLM is trained solely on unlabeled data using self-supervised objectives like masked language modeling, without any supervised fine-tuning.",,A large language model that is trained solely on unlabeled data using self-supervised objectives like masked language modeling without any supervised fine-tuning.,A large language model that is trained solely on unlabeled data using self-supervised objectives like masked language modeling without any supervised fine-tuning.,,,,AIO:LargeLanguageModel +AIO:OrdinalLLM,AIO:ModelSubset,Ordinal Large Language Model,ranking|preference modeling,Ordinal LLM,FALSE,0.33,"An ordinal LLM is trained to model ordinal relationships and rank outputs, rather than model probability distributions over text sequences directly.",,A large language model that is trained to model ordinal relationships and rank outputs rather than model probability distributions over text sequences directly.,A large language model that is trained to model ordinal relationships and rank outputs rather than model probability distributions over text sequences directly.,,,,AIO:LargeLanguageModel +AIO:ContrastiveLearningLLM,AIO:ModelSubset,,Representation learning,Contrastive Learning LLM,FALSE,0.25,"A contrastive learning LLM is trained to pull semantically similar samples closer together and push dissimilar samples apart in the representation space, learning high-quality features useful for downstream tasks.",,A large language model that is trained to pull semantically similar samples closer together and push dissimilar samples apart in the representation space learning high-quality features useful for downstream tasks.,A large language model that is trained to pull semantically similar samples closer together and push dissimilar samples apart in the representation space learning high-quality features useful for downstream tasks.,,,,AIO:LargeLanguageModel +AIO:GenerativeCommonsenseLLM,AIO:ModelSubset,Generative Commonsense Large Language Model|World Model,physical reasoning|causal modeling,Generative Commonsense LLM,FALSE,0.38,"A generative commonsense LLM is trained to understand and model basic physics, causality, and common sense about how the real world works.",,A large language model that is trained to understand and model basic physics causality and common sense about how the real world works.,A large language model that is trained to understand and model basic physics causality and common sense about how the real world works.,https://arxiv.org/abs/2306.12672,,,AIO:LargeLanguageModel +AIO:CompositionalGeneralizationLLM,AIO:ModelSubset,Compositional Generalization Large Language Model,systematic generalization|out-of-distribution generalization,Compositional Generalization LLM,FALSE,0.24,"A compositional generalization LLM is trained to understand and recombine the underlying compositional structures in language, enabling better generalization to novel combinations and out-of-distribution examples.",,A large language model that is trained to understand and recombine the underlying compositional structures in language enabling better generalization to novel combinations and out-of-distribution examples.,A large language model that is trained to understand and recombine the underlying compositional structures in language enabling better generalization to novel combinations and out-of-distribution examples.,,,,AIO:LargeLanguageModel +AIO:EthicalLLM,AIO:ModelSubset,Ethical Large Language Model,value alignment|constituitional AI,Ethical LLM,FALSE,0.38,"An ethical LLM is trained to uphold certain ethical principles, values, or rules in its language generation to increase safety and trustworthiness.",,A large language model that is trained to uphold certain ethical principles values or rules in its language generation to increase safety and trustworthiness.,A large language model that is trained to uphold certain ethical principles values or rules in its language generation to increase safety and trustworthiness.,,,,AIO:LargeLanguageModel +AIO:MultimodalFusionLLM,AIO:ModelSubset,,cross-modal grounding,Multimodal Fusion LLM,FALSE,0.37,"A multimodal fusion LLM learns joint representations across different modalities like text, vision, and audio in an end-to-end fashion for better cross-modal understanding and generation.",,A large language model that learns joint representations across different modalities like text vision and audio in an end-to-end fashion for better cross-modal understanding and generation.,A large language model that learns joint representations across different modalities like text vision and audio in an end-to-end fashion for better cross-modal understanding and generation.,,,,AIO:LargeLanguageModel +AIO:TransferLearningLLM,AIO:ModelSubset,Transfer LLM,transfer learning,Transfer Learning LLM,FALSE,0.28,"A transfer learning LLM leverages knowledge acquired during training on one task to improve performance on different but related tasks, facilitating more efficient learning and adaptation.",,A large language model that leverages knowledge acquired during training on one task to improve performance on different but related tasks facilitating more efficient learning and adaptation.,A large language model that leverages knowledge acquired during training on one task to improve performance on different but related tasks facilitating more efficient learning and adaptation.,,,,AIO:LargeLanguageModel +AIO:CausalLLM,AIO:ModelSubset,Causal Large Language Model,unidirectional|autoregressive,Causal LLM,FALSE,0.30,"A causal LLM only attends to previous tokens in the sequence when generating text, modeling the probability distribution autoregressively from left-to-right or causally.",,A large language model that only attends to previous tokens in the sequence when generating text modeling the probability distribution autoregressively from left-to-right or causally.,A large language model that only attends to previous tokens in the sequence when generating text modeling the probability distribution autoregressively from left-to-right or causally.,,,,AIO:LargeLanguageModel +AIO:CrossDomainLLM,AIO:ModelSubset,Domain-General LLM,domain adaptation|cross-domain transfer,Cross-Domain LLM,FALSE,#N/A,,,"A LLM that performs well across a wide range of domains without significant loss in performance, facilitated by advanced domain adaptation techniques.","A large language model that performs well across a wide range of domains without significant loss in performance, facilitated by advanced domain adaptation techniques.",,,,AIO:LargeLanguageModel +AIO:LanguageInterfaceLLM,AIO:ModelSubset,,Interactive learning,Language Interface LLM,FALSE,0.29,"A language interface LLM supports interactive semantic parsing, enabling users to provide feedback and corrections to dynamically refine and update the language model.",,A large language model that supports interactive semantic parsing enabling users to provide feedback and corrections to dynamically refine and update the language model.,A large language model that supports interactive semantic parsing enabling users to provide feedback and corrections to dynamically refine and update the language model.,,,,AIO:LargeLanguageModel +AIO:DecoderLLM,AIO:ModelSubset,,,Decoder LLM,FALSE,0.62,"A decoder-only architecture consisting of only a decoder, trained to predict the next token in a sequence given the previous tokens. Unlike the encoder-decoder architecture, it does not have an explicit encoder and encodes information implicitly in the hidden state of the decoder, updated at each step of the generation process.","A decoder-only architecture consisting of only a decoder, trained to predict the next token in a sequence given the previous tokens. Unlike the encoder-decoder architecture, it does not have an explicit encoder and encodes information implicitly in the hidden state of the decoder, updated at each step of the generation process.",A large language model that uses a decoder-only architecture consisting of only a decoder trained to predict the next token in a sequence given the previous tokens.,A large language model that uses a decoder-only architecture consisting of only a decoder trained to predict the next token in a sequence given the previous tokens.,https://www.practicalai.io/understanding-transformer-model-architectures/#:~:text=Encoder,,,AIO:LargeLanguageModel +AIO:EncoderLLM,AIO:ModelSubset,,,Encoder LLM,FALSE,0.48,"An encoder-only architecture that encodes the input sequence into a fixed-length representation, which is then used as input to a classifier or regressor for prediction. The model has a pre-trained general-purpose encoder that requires fine-tuning for specific tasks.","An encoder-only architecture that encodes the input sequence into a fixed-length representation, which is then used as input to a classifier or regressor for prediction. The model has a pre-trained general-purpose encoder that requires fine-tuning for specific tasks.",A large language model that uses an encoder-only architecture to encode the input sequence into a fixed-length representation which is then used as input to a classifier or regressor for prediction.,A large language model that uses an encoder-only architecture to encode the input sequence into a fixed-length representation which is then used as input to a classifier or regressor for prediction.,https://www.practicalai.io/understanding-transformer-model-architectures/#:~:text=Encoder,,,AIO:LargeLanguageModel +AIO:SparseLLM,AIO:ModelSubset,Sparse Large Language Model,model compression|parameter efficiency,Sparse LLM,FALSE,0.22,"A sparse LLM uses techniques like pruning or quantization to reduce the number of non-zero parameters in the model, making it more parameter-efficient and easier to deploy on resource-constrained devices.",,A large language model that uses techniques like pruning or quantization to reduce the number of non-zero parameters in the model making it more parameter-efficient and easier to deploy on resource-constrained devices.,A large language model that uses techniques like pruning or quantization to reduce the number of non-zero parameters in the model making it more parameter-efficient and easier to deploy on resource-constrained devices.,,,,AIO:LargeLanguageModel +AIO:RetrievalAugmentedLLM,AIO:ModelSubset,Retrieval-Augmented Large Language Model,knowledge grounding|open-book question answering,Retrieval-Augmented LLM,FALSE,#N/A,,,"A LLM which combines a pre-trained language model with a retrieval system that can access external knowledge sources. This allows the model to condition its generation on relevant retrieved knowledge, improving factual accuracy and knowledge grounding.","A large language model which combines a pre-trained language model with a retrieval system that can access external knowledge sources. This allows the model to condition its generation on relevant retrieved knowledge, improving factual accuracy and knowledge grounding.",,,,AIO:LargeLanguageModel +AIO:NeuroSymbolicLLM,AIO:ModelSubset,Neuro-Symbolic Large Language Model,knowledge reasoning|symbolic grounding,Neuro-Symbolic LLM,FALSE,#N/A,,,"A LLM which combines neural language modeling with symbolic reasoning components, leveraging structured knowledge representations and logical inferences to improve reasoning capabilities.","A large language model which combines neural language modeling with symbolic reasoning components, leveraging structured knowledge representations and logical inferences to improve reasoning capabilities.",,,,AIO:LargeLanguageModel +AIO:SemiSupervisedLLM,AIO:ModelSubset,Semi-Supervised Large Language Model,self-training,Semi-Supervised LLM,FALSE,#N/A,,,A LLM which combines self-supervised pretraining on unlabeled data with supervised fine-tuning on labeled task data.,A large language model which combines self-supervised pretraining on unlabeled data with supervised fine-tuning on labeled task data.,,,,AIO:LargeLanguageModel +AIO:MixtureofExpertsLLM,AIO:ModelSubset,MoE Large Language Model|Mixture-of-Experts Large Language Model,conditional computation|model parallelism,Mixture-of-Experts LLM,FALSE,#N/A,,,"A LLM which dynamically selects and combines outputs from multiple expert submodels, allowing for efficient scaling by conditionally activating only a subset of model components for each input.","A large language model which dynamically selects and combines outputs from multiple expert submodels, allowing for efficient scaling by conditionally activating only a subset of model components for each input.",https://proceedings.mlr.press/v162/du22c.html,,,AIO:LargeLanguageModel +AIO:GenerativeAdversarialNetworkAugmentedLLM,AIO:ModelSubset,GAN-Large Language Model|Generative Adversarial Network-Augmented Large Language Model,text generation|adversarial training,Generative Adversarial Network-Augmented LLM,FALSE,#N/A,,,"A LLM which incorporates a generative adversarial network (GAN) into its training process, using a discriminator network to provide a signal for generating more realistic and coherent text. This adversarial training can improve the quality and diversity of generated text.","A large language model which incorporates a generative adversarial network (GAN) into its training process, using a discriminator network to provide a signal for generating more realistic and coherent text. This adversarial training can improve the quality and diversity of generated text.",,,,AIO:LargeLanguageModel +AIO:KnowledgeGroundedLLM,AIO:ModelSubset,Knowledge-Grounded Large Language Model,factual grounding|knowledge integration,Knowledge-Grounded LLM,FALSE,#N/A,,,"A LLM which incorporates external knowledge sources or knowledge bases into the model architecture, enabling it to generate more factually accurate and knowledge-aware text.","A large language model which incorporates external knowledge sources or knowledge bases into the model architecture, enabling it to generate more factually accurate and knowledge-aware text.",,,,AIO:LargeLanguageModel +AIO:MemoryAugmentedLLM,AIO:ModelSubset,Memory-Augmented Large Language Model,external memory,Memory-Augmented LLM,FALSE,#N/A,,,"A LLM which incorporates external writable and readable memory components, allowing it to store and retrieve information over long contexts.","A large language model which incorporates external writable and readable memory components, allowing it to store and retrieve information over long contexts.",https://arxiv.org/abs/2306.07174,,,AIO:LargeLanguageModel +AIO:PromptbasedFineTuningLLM,AIO:ModelSubset,Prompt-tuned Large Language Model|Prompt-based Fine-Tuning Large Language Model,in-context learning|few-shot learning,Prompt-based Fine-Tuning LLM,FALSE,#N/A,,,"A LLM which is fine-tuned on a small number of examples or prompts, rather than full task datasets. This allows for rapid adaptation to new tasks with limited data, leveraging the model's few-shot learning capabilities.","A large language model which is fine-tuned on a small number of examples or prompts, rather than full task datasets. This allows for rapid adaptation to new tasks with limited data, leveraging the model's few-shot learning capabilities.",,,,AIO:LargeLanguageModel +AIO:InstructionTunedLLM,AIO:ModelSubset,Instruction-Tuned Large Language Model,natural language instructions|constitutional AI,Instruction-Tuned LLM,FALSE,#N/A,,,"A LLM which is fine-tuned to follow natural language instructions accurately and safely, learning to map from instructions to desired model behavior in a more controlled and principled way.","A large language model which is fine-tuned to follow natural language instructions accurately and safely, learning to map from instructions to desired model behavior in a more controlled and principled way.",,,,AIO:LargeLanguageModel +AIO:LowResourceLLM,AIO:ModelSubset,Low-Resource Language Model,resource-efficient|low-resource languages,Low-Resource LLM,FALSE,#N/A,,,"A LLM which is optimized for performance in scenarios with limited data, computational resources, or for languages with sparse datasets.","A large language model which is optimized for performance in scenarios with limited data, computational resources, or for languages with sparse datasets.",,,,AIO:LargeLanguageModel +AIO:DomainAdaptedLLM,AIO:ModelSubset,Domain-Adapted Large Language Model,transfer learning|domain robustness,Domain-Adapted LLM,FALSE,#N/A,,,"A LLM which is pre-trained on a broad corpus and then fine-tuned on domain-specific data to specialize its capabilities for particular domains or applications, like scientific literature or code generation.","A large language model which is pre-trained on a broad corpus and then fine-tuned on domain-specific data to specialize its capabilities for particular domains or applications, like scientific literature or code generation.",,,,AIO:LargeLanguageModel +AIO:MetaLearningLLM,AIO:ModelSubset,Meta-Learning Large Language Model,few-shot adaptation|learning to learn,Meta-Learning LLM,FALSE,#N/A,,,"A LLM which is trained in a way that allows it to quickly adapt to new tasks or datasets through only a few examples or fine-tuning steps, leveraging meta-learned priors about how to efficiently learn.","A large language model which is trained in a way that allows it to quickly adapt to new tasks or datasets through only a few examples or fine-tuning steps, leveraging meta-learned priors about how to efficiently learn.",,,,AIO:LargeLanguageModel +AIO:MultiTaskLLM,AIO:ModelSubset,Multi-Task Large Language Model,transfer learning,Multi-Task LLM,FALSE,#N/A,,,"A LLM which is trained jointly on multiple language tasks simultaneously, learning shared representations that transfer across tasks.","A large language model which is trained jointly on multiple language tasks simultaneously, learning shared representations that transfer across tasks.",,,,AIO:LargeLanguageModel +AIO:SelfSupervisedLLM,AIO:ModelSubset,,Pretext tasks,Self-Supervised LLM,FALSE,#N/A,,,"A LLM which learns rich representations by solving pretext tasks that involve predicting parts of the input from other observed parts of the data, without relying on human-annotated labels.","A large language model which learns rich representations by solving pretext tasks that involve predicting parts of the input from other observed parts of the data, without relying on human-annotated labels.",,,,AIO:LargeLanguageModel +AIO:EnergyBasedLLM,AIO:ModelSubset,Energy-Based Large Language Model,energy scoring|explicit density modeling,Energy-Based LLM,FALSE,#N/A,,,"A LLM which models the explicit probability density over token sequences using an energy function, rather than an autoregressive factorization. This can improve modeling of long-range dependencies and global coherence.","A large language model which models the explicit probability density over token sequences using an energy function, rather than an autoregressive factorization. This can improve modeling of long-range dependencies and global coherence.",,,,AIO:LargeLanguageModel +AIO:ZeroShotLearningLLM,AIO:ModelSubset,Zero-Shot LLM,zero-shot learning,Zero-Shot Learning LLM,FALSE,#N/A,,,"A LLM which performs tasks or understands concepts it has not explicitly been trained on, demonstrating a high degree of generalization and understanding.","A large language model which performs tasks or understands concepts it has not explicitly been trained on, demonstrating a high degree of generalization and understanding.",,,,AIO:LargeLanguageModel +AIO:InputLayer,AIO:LayerSubset,,,Input Layer,FALSE,0.52,"The input layer of a neural network is composed of artificial input neurons, and brings the initial data into the system for further processing by subsequent layers of artificial neurons. The input layer is the very beginning of the workflow for the artificial neural network.","The input layer of a neural network is composed of artificial input neurons, and brings the initial data into the system for further processing by subsequent layers of artificial neurons. The input layer is the very beginning of the workflow for the artificial neural network.",A layer composed of artificial input neurons that brings the initial data into the system for further processing by subsequent layers.,A layer composed of artificial input neurons that brings the initial data into the system for further processing by subsequent layers.,https://www.techopedia.com/definition/33262/input-layer-neural-networks,,,AIO:Layer +AIO:RecurrentLayer,AIO:LayerSubset,,,Recurrent Layer,FALSE,0.37,"A layer of an RNB, composed of recurrent units and with the number of which is the hidden size of the layer.",,A layer composed of recurrent units with the number equal to the hidden size of the layer.,A layer composed of recurrent units with the number equal to the hidden size of the layer.,https://docs.nvidia.com/deepLearning/performance/dl-performance-recurrent/index.html#recurrent-layer,,,AIO:Layer +AIO:OutputLayer,AIO:LayerSubset,,,Output Layer,FALSE,0.71,"The output layer in an artificial neural network is the last layer of neurons that produces given outputs for the program. Though they are made much like other artificial neurons in the neural network, output layer neurons may be built or observed in a different way, given that they are the last “actor” nodes on the network.","The output layer in an artificial neural network is the last layer of neurons that produces given outputs for the program. Though they are made much like other artificial neurons in the neural network, output layer neurons may be built or observed in a different way, given that they are the last “actor” nodes on the network.",A layer containing the last neurons in the network that produces given outputs for the program.,A layer containing the last neurons in the network that produces given outputs for the program.,https://www.techopedia.com/definition/33263/output-layer-neural-networks,,,AIO:Layer +AIO:HiddenLayer,AIO:LayerSubset,,,Hidden Layer,FALSE,0.69,"A hidden layer is located between the input and output of the algorithm, in which the function applies weights to the inputs and directs them through an activation function as the output. In short, the hidden layers perform nonlinear transformations of the inputs entered into the network. Hidden layers vary depending on the function of the neural network, and similarly, the layers may vary depending on their associated weights.","A hidden layer is located between the input and output of the algorithm, in which the function applies weights to the inputs and directs them through an activation function as the output. In short, the hidden layers perform nonlinear transformations of the inputs entered into the network. Hidden layers vary depending on the function of the neural network, and similarly, the layers may vary depending on their associated weights.",A layer located between the input and output that performs nonlinear transformations of the inputs entered into the network.,A layer located between the input and output that performs nonlinear transformations of the inputs entered into the network.,https://deepai.org/machine-Learning-glossary-and-terms/hidden-layer-machine-Learning,,,AIO:Layer +AIO:MemoryCellLayer,AIO:LayerSubset,,,Memory Cell Layer,FALSE,#N/A,,,"A layer of cells, each with an internal state or weights.","A layer of cells, each with an internal state or weights.",https://doi.org/10.1162/neco.1997.9.8.1735,,,AIO:Layer +AIO:WeightedLayer,AIO:LayerSubset,,,Weighted Layer,FALSE,#N/A,,,A layer of values to be applied to other cells or neurons in a network.,A layer of values to be applied to other cells or neurons in a network.,,,,AIO:Layer +AIO:AbstractRNNCell,AIO:LayerSubset,,,AbstractRNNCell,FALSE,0.36,Abstract object representing an RNN cell. This is the base class for implementing RNN cells with custom behavior.,,An abstract layer object representing an RNN cell that is the base class for implementing RNN cells with custom behavior.,A layer representing an RNN cell that is the base class for implementing RNN cells with custom behavior.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/AbstractRNNCell,,,AIO:Layer +AIO:AdditionLayer,AIO:LayerSubset,,,Addition Layer,FALSE,#N/A,,,A layer that adds inputs from one or more other layers to cells or neurons of a target layer.,A layer that adds inputs from one or more other layers to cells or neurons of a target layer.,,,,AIO:Layer +AIO:StackedRNNCellsLayer,AIO:LayerSubset,,,StackedRNNCells Layer,FALSE,0.52,Wrapper allowing a stack of RNN cells to behave as a single cell. Used to implement efficient stacked RNNs.,Wrapper allowing a stack of RNN cells to behave as a single cell. Used to implement efficient stacked RNNs.,A layer that allows a stack of RNN cells to behave as a single cell.,A layer that allows a stack of RNN cells to behave as a single cell.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/StackedRNNCells,,,AIO:Layer +AIO:ActivationLayer,AIO:LayerSubset,,,Activation Layer,FALSE,0.50,Applies an activation function to an output.,Applies an activation function to an output.,A layer that applies an activation function to an output.,A layer that applies an activation function to an output.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Activation,,,AIO:Layer +AIO:RegularizationLayer,AIO:LayerSubset,,,Regularization Layer,FALSE,0.52,Regularizers allow you to apply penalties on layer parameters or layer activity during optimization. These penalties are summed into the loss function that the network optimizes. Regularization penalties are applied on a per-layer basis.,Regularizers allow you to apply penalties on layer parameters or layer activity during optimization. These penalties are summed into the loss function that the network optimizes. Regularization penalties are applied on a per-layer basis.,A layer that applies penalties on layer parameters or layer activity during optimization summed into the loss function that the network optimizes.,A layer that applies penalties on layer parameters or layer activity during optimization summed into the loss function that the network optimizes.,https://keras.io/api/layers/regularizers/,,,AIO:Layer +AIO:WrapperLayer,AIO:LayerSubset,,,Wrapper Layer,FALSE,0.83,"Abstract wrapper base class. Wrappers take another layer and augment it in various ways. Do not use this class as a layer, it is only an abstract base class. Two usable wrappers are the TimeDistributed and Bidirectional wrappers.","Abstract wrapper base class. Wrappers take another layer and augment it in various ways. Do not use this class as a layer, it is only an abstract base class. Two usable wrappers are the TimeDistributed and Bidirectional wrappers.",An abstract base class for wrappers that augment the functionality of another layer.,A layer that augment the functionality of another layer.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Wrapper,,,AIO:Layer +AIO:DotLayer,AIO:LayerSubset,,,Dot Layer,FALSE,0.73,"Layer that computes a dot product between samples in two tensors. E.g. if applied to a list of two tensors a and b of shape (batch_size, n), the output will be a tensor of shape (batch_size, 1) where each entry i will be the dot product between a[i] and b[i].","Layer that computes a dot product between samples in two tensors. E.g. if applied to a list of two tensors a and b of shape (batch_size, n), the output will be a tensor of shape (batch_size, 1) where each entry i will be the dot product between a[i] and b[i].",A layer that computes a dot product between samples in two tensors.,A layer that computes a dot product between samples in two tensors.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Dot,,,AIO:Layer +AIO:ConvolutionalLayer,AIO:LayerSubset,,,Convolutional Layer,FALSE,0.61,"A convolutional layer is the main building block of a CNN. It contains a set of filters (or kernels), parameters of which are to be learned throughout the training. The size of the filters is usually smaller than the actual image. Each filter convolves with the image and creates an activation map.","A convolutional layer is the main building block of a CNN. It contains a set of filters (or kernels), parameters of which are to be learned throughout the training. The size of the filters is usually smaller than the actual image. Each filter convolves with the image and creates an activation map.",A layer that contains a set of filters (or kernels) parameters of which are to be learned throughout the training.,A layer that contains a set of filters (or kernels) parameters of which are to be learned throughout the training.,https://www.sciencedirect.com/topics/engineering/convolutional-layer,,,AIO:Layer +AIO:Cropping3DLayer,AIO:LayerSubset,,,Cropping3D Layer,FALSE,0.87,Cropping layer for 3D data (e.g. spatial or spatio-temporal).,Cropping layer for 3D data (e.g. spatial or spatio-temporal).,A layer that crops along spatial dimensions (depth,A layer that crops along spatial dimensions (depth,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Cropping3D,,,AIO:Layer +AIO:Cropping2DLayer,AIO:LayerSubset,,,Cropping2D Layer,FALSE,0.64,"Cropping layer for 2D input (e.g. picture). It crops along spatial dimensions, i.e. height and width.","Cropping layer for 2D input (e.g. picture). It crops along spatial dimensions, i.e. height and width.",A layer that crops along spatial dimensions (i.e. height and width) for 2D input.,A layer that crops along spatial dimensions (i.e. height and width) for 2D input.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Cropping2D,,,AIO:Layer +AIO:LayerLayer,AIO:LayerSubset,,,Layer Layer,FALSE,0.93,"This is the class from which all layers inherit. A layer is a callable object that takes as input one or more tensors and that outputs one or more tensors. It involves computation, defined in the call() method, and a state (weight variables). State can be created in various places, at the convenience of the subclass implementer: in __init__(); in the optional build() method, which is invoked by the first __call__() to the layer, and supplies the shape(s) of the input(s), which may not have been known at initialization time; in the first invocation of call(), with some caveats discussed below. Users will just instantiate a layer and then treat it as a callable.","This is the class from which all layers inherit. A layer is a callable object that takes as input one or more tensors and that outputs one or more tensors. It involves computation, defined in the call() method, and a state (weight variables). State can be created in various places, at the convenience of the subclass implementer: in __init__(); in the optional build() method, which is invoked by the first __call__() to the layer, and supplies the shape(s) of the input(s), which may not have been known at initialization time; in the first invocation of call(), with some caveats discussed below. Users will just instantiate a layer and then treat it as a callable.",The base class from which all layers inherit.,A layer that form which other layers can inherit.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Layer,,,AIO:Layer +AIO:AttentionLayer,AIO:LayerSubset,,,Attention Layer,FALSE,0.79,"Dot-product attention layer, a.k.a. Luong-style attention.","Dot-product attention layer, a.k.a. Luong-style attention.",A layer that implements dot-product attention also known as Luong-style attention.,A layer that implements dot-product attention also known as Luong-style attention.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Attention,,,AIO:Layer +AIO:NoiseDenseLayer,AIO:LayerSubset,,,Noise Dense Layer,FALSE,0.90,"Noisy dense layer that injects random noise to the weights of dense layer. Noisy dense layers are fully connected layers whose weights and biases are augmented by factorised Gaussian noise. The factorised Gaussian noise is controlled through gradient descent by a second weights layer. A NoisyDense layer implements the operation: $$ \mathrm{NoisyDense}(x) = \mathrm{activation}(\mathrm{dot}(x, \mu + (\sigma \cdot \epsilon)) \mathrm{bias}) $$ where mu is the standard weights layer, epsilon is the factorised Gaussian noise, and delta is a second weights layer which controls epsilon.","Noisy dense layer that injects random noise to the weights of dense layer. Noisy dense layers are fully connected layers whose weights and biases are augmented by factorised Gaussian noise. The factorised Gaussian noise is controlled through gradient descent by a second weights layer. A NoisyDense layer implements the operation: $$ \mathrm{NoisyDense}(x) = \mathrm{activation}(\mathrm{dot}(x, \mu + (\sigma \cdot \epsilon)) \mathrm{bias}) $$ where mu is the standard weights layer, epsilon is the factorised Gaussian noise, and delta is a second weights layer which controls epsilon.",A layer that is a densely-connected neural network layer with added noise for regularization.,A layer that is a densely-connected neural network layer with added noise for regularization.,https://www.tensorflow.org/addons/api_docs/python/tfa/layers/NoisyDense,,,AIO:Layer +AIO:DenseLayer,AIO:LayerSubset,,,Dense Layer,FALSE,0.77,Just your regular densely-connected NN layer.,Just your regular densely-connected NN layer.,A layer that is a regular densely-connected neural network layer.,A layer that is a regular densely-connected neural network layer.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Dense,,,AIO:Layer +AIO:ReshapingLayer,AIO:LayerSubset,Reshape Layer,,Reshaping Layer,FALSE,0.50,Reshape layers are used to change the shape of the input.,Reshape layers are used to change the shape of the input.,A layer that is used to change the shape of the input.,A layer that is used to change the shape of the input.,https://keras.io/api/layers/reshaping_layers/reshape/,,,AIO:Layer +AIO:MaskingLayer,AIO:LayerSubset,,,Masking Layer,FALSE,0.80,"Masks a sequence by using a mask value to skip timesteps. For each timestep in the input tensor (dimension #1 in the tensor), if all values in the input tensor at that timestep are equal to mask_value, then the timestep will be masked (skipped) in all downstream layers (as long as they support masking). If any downstream layer does not support masking yet receives such an input mask, an exception will be raised.","Masks a sequence by using a mask value to skip timesteps. For each timestep in the input tensor (dimension #1 in the tensor), if all values in the input tensor at that timestep are equal to mask_value, then the timestep will be masked (skipped) in all downstream layers (as long as they support masking). If any downstream layer does not support masking yet receives such an input mask, an exception will be raised.",A layer that masks a sequence by using a mask value to skip timesteps.,A layer that masks a sequence by using a mask value to skip timesteps.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Masking,,,AIO:Layer +AIO:KernelLayer,AIO:LayerSubset,,,Kernel Layer,FALSE,#N/A,,,A layer that obtains the dot product of input values or subsets of input values.,A layer that obtains the dot product of input values or subsets of input values.,,,,AIO:Layer +AIO:CategoricalFeaturesPreprocessingLayer,AIO:LayerSubset,,,Categorical Features Preprocessing Layer,FALSE,0.00,A layer that performs categorical data preprocessing operations.,,A layer that performs categorical data preprocessing operations.,A layer that performs categorical data preprocessing operations.,https://keras.io/guides/preprocessing_layers/,,,AIO:Layer +AIO:PreprocessingLayer,AIO:LayerSubset,,,Preprocessing Layer,FALSE,0.00,A layer that performs data preprocessing operations.,,A layer that performs data preprocessing operations.,A layer that performs data preprocessing operations.,https://www.tensorflow.org/guide/keras/preprocessing_layers,,,AIO:Layer +AIO:ImageAugmentationLayer,AIO:LayerSubset,,,Image Augmentation Layer,FALSE,0.00,A layer that performs image data preprocessing augmentations.,,A layer that performs image data preprocessing augmentations.,A layer that performs image data preprocessing augmentations.,https://keras.io/guides/preprocessing_layers/,,,AIO:Layer +AIO:ImagePreprocessingLayer,AIO:LayerSubset,,,Image Preprocessing Layer,FALSE,0.00,A layer that performs image data preprocessing operations.,,A layer that performs image data preprocessing operations.,A layer that performs image data preprocessing operations.,https://keras.io/guides/preprocessing_layers/,,,AIO:Layer +AIO:NumericalFeaturesPreprocessingLayer,AIO:LayerSubset,,,Numerical Features Preprocessing Layer,FALSE,0.00,A layer that performs numerical data preprocessing operations.,,A layer that performs numerical data preprocessing operations.,A layer that performs numerical data preprocessing operations.,https://keras.io/guides/preprocessing_layers/,,,AIO:Layer +AIO:TextPreprocessingLayer,AIO:LayerSubset,,,Text Preprocessing Layer,FALSE,0.00,A layer that performs text data preprocessing operations.,,A layer that performs text data preprocessing operations.,A layer that performs text data preprocessing operations.,https://keras.io/guides/preprocessing_layers/,,,AIO:Layer +AIO:GRUCellLayer,AIO:LayerSubset,,,GRUCell Layer,FALSE,0.52,"Cell class for the GRU layer. This class processes one step within the whole time sequence input, whereas tf.keras.layer.GRU processes the whole sequence.","Cell class for the GRU layer. This class processes one step within the whole time sequence input, whereas tf.keras.layer.GRU processes the whole sequence.",A layer that processes one step within the whole time sequence input for a GRU layer.,A layer that processes one step within the whole time sequence input for a GRU layer.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GRUCell,,,AIO:Layer +AIO:SimpleRNNCellLayer,AIO:LayerSubset,,,SimpleRNNCell Layer,FALSE,0.63,"Cell class for SimpleRNN. This class processes one step within the whole time sequence input, whereas tf.keras.layer.SimpleRNN processes the whole sequence.","Cell class for SimpleRNN. This class processes one step within the whole time sequence input, whereas tf.keras.layer.SimpleRNN processes the whole sequence.",A layer that processes one step within the whole time sequence input for a SimpleRNN layer.,A layer that processes one step within the whole time sequence input for a SimpleRNN layer.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/SimpleRNNCell,,,AIO:Layer +AIO:LSTMCellLayer,AIO:LayerSubset,,,LSTMCell Layer,FALSE,0.78,Cell class for the LSTM layer.,Cell class for the LSTM layer.,A layer that processes one step within the whole time sequence input for an LSTM layer.,A layer that processes one step within the whole time sequence input for an LSTM layer.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/LSTMCell,,,AIO:Layer +AIO:DenseFeaturesLayer,AIO:LayerSubset,,,DenseFeatures Layer,FALSE,0.86,"A layer that produces a dense Tensor based on given feature_columns. Generally a single example in training data is described with FeatureColumns. At the first layer of the model, this column oriented data should be converted to a single Tensor. This layer can be called multiple times with different features. This is the V2 version of this layer that uses name_scopes to create variables instead of variable_scopes. But this approach currently lacks support for partitioned variables. In that case, use the V1 version instead.","A layer that produces a dense Tensor based on given feature_columns. Generally a single example in training data is described with FeatureColumns. At the first layer of the model, this column oriented data should be converted to a single Tensor. This layer can be called multiple times with different features. This is the V2 version of this layer that uses name_scopes to create variables instead of variable_scopes. But this approach currently lacks support for partitioned variables. In that case, use the V1 version instead.",A layer that produces a dense tensor based on given feature columns.,A layer that produces a dense tensor based on given feature columns.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/DenseFeatures,,,AIO:Layer +AIO:PoolingLayer,AIO:LayerSubset,,,Pooling Layer,FALSE,0.55,Pooling layers serve the dual purposes of mitigating the sensitivity of convolutional layers to location and of spatially downsampling representations.,Pooling layers serve the dual purposes of mitigating the sensitivity of convolutional layers to location and of spatially downsampling representations.,A layer that serves to mitigate the sensitivity of convolutional layers to location and spatially downsample representations.,A layer that serves to mitigate the sensitivity of convolutional layers to location and spatially downsample representations.,https://d2l.ai/chapter_convolutional-neural-networks/pooling.html,,,AIO:Layer +AIO:InputSpecLayer,AIO:LayerSubset,,,InputSpec Layer,FALSE,0.94,"Specifies the rank, dtype and shape of every input to a layer. Layers can expose (if appropriate) an input_spec attribute: an instance of InputSpec, or a nested structure of InputSpec instances (one per input tensor). These objects enable the layer to run input compatibility checks for input structure, input rank, input shape, and input dtype. A None entry in a shape is compatible with any dimension, a None shape is compatible with any shape.","Specifies the rank, dtype and shape of every input to a layer. Layers can expose (if appropriate) an input_spec attribute: an instance of InputSpec, or a nested structure of InputSpec instances (one per input tensor). These objects enable the layer to run input compatibility checks for input structure, input rank, input shape, and input dtype. A None entry in a shape is compatible with any dimension, a None shape is compatible with any shape.",A layer that specifies the rank,A layer that specifies the rank,https://www.tensorflow.org/api_docs/python/tf/keras/layers/InputSpec,,,AIO:Layer +AIO:EmbeddingLayer,AIO:LayerSubset,,,Embedding Layer,FALSE,0.36,Turns positive integers (indexes) into dense vectors of fixed size.,,A layer that turns positive integers (indexes) into dense vectors of fixed size.,A layer that turns positive integers (indexes) into dense vectors of fixed size.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Embedding,,,AIO:Layer +AIO:UpSampling3DLayer,AIO:LayerSubset,,,UpSampling3D Layer,FALSE,0.93,Upsampling layer for 3D inputs.,Upsampling layer for 3D inputs.,A layer that upsamples the input by repeating each depth,A layer that upsamples the input by repeating each depth,https://www.tensorflow.org/api_docs/python/tf/keras/layers/UpSampling3D,,,AIO:Layer +AIO:UpSampling2DLayer,AIO:LayerSubset,,,UpSampling2D Layer,FALSE,0.85,Upsampling layer for 2D inputs. Repeats the rows and columns of the data by size[0] and size[1] respectively.,Upsampling layer for 2D inputs. Repeats the rows and columns of the data by size[0] and size[1] respectively.,A layer that upsamples the input by repeating each row and column size times.,A layer that upsamples the input by repeating each row and column size times.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/UpSampling2D,,,AIO:Layer +AIO:LocallyconnectedLayer,AIO:LayerSubset,,,Locally-connected Layer,FALSE,#N/A,,,"A layer that works similarly to the Convolution1D layer, except that weights are unshared, that is, a different set of filters is applied at each different patch of the input.","A layer that works similarly to the Convolution1D layer, except that weights are unshared, that is, a different set of filters is applied at each different patch of the input.",https://faroit.com/keras-docs/1.2.2/layers/local/,,,AIO:Layer +AIO:LambdaLayer,AIO:LayerSubset,,,Lambda Layer,FALSE,0.45,Wraps arbitrary expressions as a Layer object.,Wraps arbitrary expressions as a Layer object.,A layer that wraps arbitrary expressions as a Layer object.,A layer that wraps arbitrary expressions as a Layer object.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Lambda,,,AIO:Layer +AIO:PolicyLayer,AIO:LayerSubset,,,Policy Layer,FALSE,#N/A,,,"A layer that, after taking a set of states or values as input, predicts a probability distribution of actions to take.","A layer that, after taking a set of states or values as input, predicts a probability distribution of actions to take.",,,,AIO:Layer +AIO:InputLayerLayer,AIO:LayerSubset,,,InputLayer Layer,FALSE,0.18,Layer to be used as an entry point into a Network (a graph of layers).,,A layer to be used as an entry point into a Network (a graph of layers).,A layer to be used as an entry point into a Network (a graph of layers).,https://www.tensorflow.org/api_docs/python/tf/keras/layers/InputLayer,,,AIO:Layer +AIO:MergingLayer,AIO:LayerSubset,,,Merging Layer,FALSE,0.00,A layer used to merge a list of inputs.,,A layer used to merge a list of inputs.,A layer used to merge a list of inputs.,https://www.tutorialspoint.com/keras/keras_merge_layer.htm,,,AIO:Layer +AIO:LocallyConnected1DLayer,AIO:LayerSubset,,,LocallyConnected1D Layer,FALSE,0.71,"Locally-connected layer for 1D inputs. The LocallyConnected1D layer works similarly to the Conv1D layer, except that weights are unshared, that is, a different set of filters is applied at each different patch of the input.","Locally-connected layer for 1D inputs. The LocallyConnected1D layer works similarly to the Conv1D layer, except that weights are unshared, that is, a different set of filters is applied at each different patch of the input.",A locally-connected layer for 1D inputs where each patch of the input is convolved with a different set of filters.,A locally-connected layer for 1D inputs where each patch of the input is convolved with a different set of filters.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/LocallyConnected1D,,,AIO:LocallyconnectedLayer +AIO:LocallyConnected2DLayer,AIO:LayerSubset,,,LocallyConnected2D Layer,FALSE,0.71,"Locally-connected layer for 2D inputs. The LocallyConnected2D layer works similarly to the Conv2D layer, except that weights are unshared, that is, a different set of filters is applied at each different patch of the input.","Locally-connected layer for 2D inputs. The LocallyConnected2D layer works similarly to the Conv2D layer, except that weights are unshared, that is, a different set of filters is applied at each different patch of the input.",A locally-connected layer for 2D inputs where each patch of the input is convolved with a different set of filters.,A locally-connected layer for 2D inputs where each patch of the input is convolved with a different set of filters.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/LocallyConnected2D,,,AIO:LocallyconnectedLayer +AIO:GatedRecurrentUnit,AIO:NetworkSubset,GRU,,Gated Recurrent Unit,FALSE,0.43,"Gated recurrent units (GRUs) are a gating mechanism in recurrent neural networks, similar to LSTMs but with fewer parameters and no output gate.",,A long short-term memory network that is a gating mechanism in recurrent neural networks similar to LSTMs but with fewer parameters and no output gate.,A long short-term memory network that is a gating mechanism in recurrent neural networks similar to LSTMs but with fewer parameters and no output gate.,https://en.wikipedia.org/wiki/Gated_recurrent_unit,"Layers: Input, Memory Cell, Output",AIO:InputLayer|AIO:MemoryCellLayer|AIO:OutputLayer,AIO:LongShortTermMemory +AIO:node2vec,AIO:NetworkSubset,N2V,,node2vec,FALSE,#N/A,,,A machine learning designed to learn continuous feature representations for nodes in a graph by optimizing a neighborhood-preserving objective.,A machine learning task designed to learn continuous feature representations for nodes in a graph by optimizing a neighborhood-preserving objective.,https://en.wikipedia.org/wiki/Node2vec,"Layers: Input, Hidden, Output",AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:MachineLearningTask +AIO:UnsupervisedLearning,AIO:MachineLearningSubset,,,Unsupervised Learning,FALSE,0.60,Algorithms that learn patterns from unlabeled data.,Algorithms that learn patterns from unlabeled data.,A type of machine learning focused on algorithms that learn patterns from unlabeled data.,A machine learning task focused on algorithms that learn patterns from unlabeled data.,https://en.wikipedia.org/wiki/Unsupervised_learning,,,AIO:MachineLearningTask +AIO:SurvivalAnalysis,AIO:MachineLearningSubset,,,Survival Analysis,FALSE,0.32,"Methods for analyzing the expected duration of time until one or more events occur, such as death in biological organisms or failure in mechanical systems.",,A machine learning task focused on methods for analyzing the expected duration of time until one or more events occur such as death in biological organisms or failure in mechanical systems.,A machine learning task focused on methods for analyzing the expected duration of time until one or more events occur such as death in biological organisms or failure in mechanical systems.,https://en.wikipedia.org/wiki/Survival_analysis,,,AIO:MachineLearningTask +AIO:TimeSeriesAnalysis,AIO:MachineLearningSubset,,,Time Series Analysis,FALSE,0.42,Methods for analyzing time series data to extract meaningful statistics and characteristics.,Methods for analyzing time series data to extract meaningful statistics and characteristics.,A machine learning task focused on methods for analyzing time series data to extract meaningful statistics and characteristics.,A machine learning task focused on methods for analyzing time series data to extract meaningful statistics and characteristics.,https://en.wikipedia.org/wiki/Time_series,,,AIO:MachineLearningTask +AIO:ReinforcementLearning,AIO:MachineLearningSubset,,,Reinforcement Learning,FALSE,0.27,"Methods that do not require labeled input/output pairs or explicit correction of sub-optimal actions, focusing instead on balancing exploration and exploitation to optimize performance over time.",,A type of machine learning focused on methods that do not require labeled input/output pairs or explicit correction of sub-optimal actions focusing instead on balancing exploration and exploitation to optimize performance over time.,A machine learning task focused on methods that do not require labeled input/output pairs or explicit correction of sub-optimal actions focusing instead on balancing exploration and exploitation to optimize performance over time.,https://en.wikipedia.org/wiki/Reinforcement_learning,,,AIO:MachineLearningTask +AIO:Clustering,AIO:MachineLearningSubset,Cluster analysis,,Clustering,FALSE,0.30,Methods that group a set of objects such that objects in the same group are more similar to each other than to those in other groups.,,A machine learning task focused on methods that group a set of objects such that objects in the same group are more similar to each other than to those in other groups.,A machine learning task focused on methods that group a set of objects such that objects in the same group are more similar to each other than to those in other groups.,https://en.wikipedia.org/wiki/Cluster_analysis,,,AIO:MachineLearningTask +AIO:ActiveLearning,AIO:MachineLearningSubset,Query Learning,,Active Learning,FALSE,0.33,Methods that interactively query a user or another information source to label new data points with the desired outputs.,,A type of machine learning focused on methods that interactively query a user or another information source to label new data points with the desired outputs.,A machine learning task focused on methods that interactively query a user or another information source to label new data points with the desired outputs.,https://en.wikipedia.org/wiki/Active_learning_(machine_learning),,,AIO:MachineLearningTask +AIO:SupervisedLearning,AIO:MachineLearningSubset,,,Supervised Learning,FALSE,0.38,Methods that learn a function mapping input to output based on example input-output pairs.,,A type of machine learning focused on methods that learn a function mapping input to output based on example input-output pairs.,A machine learning task focused on methods that learn a function mapping input to output based on example input-output pairs.,https://en.wikipedia.org/wiki/Supervised_learning,,,AIO:MachineLearningTask +AIO:TimeSeriesForecasting,AIO:MachineLearningSubset,,,Time Series Forecasting,FALSE,0.44,Methods that predict future values based on previously observed values.,Methods that predict future values based on previously observed values.,A machine learning task focused on methods that predict future values based on previously observed values.,A machine learning task focused on methods that predict future values based on previously observed values.,https://en.wikipedia.org/wiki/Time_series,,,AIO:MachineLearningTask +AIO:DataImputation,AIO:MachineLearningSubset,,,Data Imputation,FALSE,0.53,Methods that replace missing data with substituted values.,Methods that replace missing data with substituted values.,A machine learning task focused on methods that replace missing data with substituted values.,A machine learning task focused on methods that replace missing data with substituted values.,https://en.wikipedia.org/wiki/Imputation_(statistics),,,AIO:MachineLearningTask +AIO:TransferLearning,AIO:NetworkSubset,,,Transfer Learning,FALSE,0.30,Methods that reuse or transfer information from previously learned tasks to facilitate the learning of new tasks.,,A type of machine learning focused on methods that reuse or transfer information from previously learned tasks to facilitate the learning of new tasks.,A machine learning task focused on methods that reuse or transfer information from previously learned tasks to facilitate the learning of new tasks.,https://en.wikipedia.org/wiki/Transfer_learning,,,AIO:MachineLearningTask +AIO:Biclustering,AIO:MachineLearningSubset,Block Clustering|Co-clustering|Two-mode Clustering|Two-way Clustering|Joint Clustering,,Biclustering,FALSE,0.33,Methods that simultaneously cluster the rows and columns of a matrix to identify submatrices with coherent patterns.,,A machine learning task focused on methods that simultaneously cluster the rows and columns of a matrix to identify submatrices with coherent patterns.,A machine learning task focused on methods that simultaneously cluster the rows and columns of a matrix to identify submatrices with coherent patterns.,https://en.wikipedia.org/wiki/Biclustering,,,AIO:MachineLearningTask +AIO:EnsembleLearning,AIO:MachineLearningSubset,,,Ensemble Learning,FALSE,0.30,Methods that use multiple learning algorithms to achieve better predictive performance than any of the constituent algorithms alone.,,A type of machine learning focused on methods that use multiple learning algorithms to achieve better predictive performance than any of the constituent algorithms alone.,A machine learning task focused on methods that use multiple learning algorithms to achieve better predictive performance than any of the constituent algorithms alone.,https://en.wikipedia.org/wiki/Ensemble_learning,,,AIO:MachineLearningTask +AIO:NaturalLanguageProcessing,AIO:MachineLearningSubset,NLP,,Natural Language Processing,FALSE,0.28,"A subfield of linguistics, computer science, and artificial intelligence focused on the interactions between computers and human language, including programming computers to process and analyze large amounts of natural language data.",,A subfield of machine learning focused on the interactions between computers and human language including programming computers to process and analyze large amounts of natural language data.,A machine learning task focused on the interactions between computers and human language including programming computers to process and analyze large amounts of natural language data.,https://en.wikipedia.org/wiki/Natural_language_processing,,,AIO:MachineLearningTask +AIO:ProbabilisticGraphicalModel,AIO:MachineLearningSubset,PGM|Graphical Model|Structure Probabilistic Model,,Probabilistic Graphical Model,FALSE,0.18,A probabilistic model in which a graph expresses the conditional dependence structure between random variables.,,A machine learning model in which a graph expresses the conditional dependence structure between random variables.,A machine learning task in which a graph expresses the conditional dependence structure between random variables.,https://en.wikipedia.org/wiki/Graphical_model,,,AIO:MachineLearningTask +AIO:MetaLearning,AIO:NetworkSubset,,,Meta-Learning,FALSE,#N/A,,,A machine learning that automatically learns from metadata about machine learning experiments.,A machine learning task that automatically learns from metadata about machine learning experiments.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:MachineLearningTask +AIO:word2vec,AIO:NetworkSubset,W2V,,word2vec,FALSE,#N/A,,,"A machine learning that generates distributed representations of words by training a shallow neural network model, which aims to predict the context of each word within a corpus. This algorithm captures semantic meanings of words through their contextual usage in the text.","A machine learning task that generates distributed representations of words by training a shallow neural network model, which aims to predict the context of each word within a corpus. This algorithm captures semantic meanings of words through their contextual usage in the text.",https://en.wikipedia.org/wiki/Word2vec,"Layers: Input, Hidden, Output",AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:MachineLearningTask +AIO:KnearestNeighborAlgorithm,AIO:MachineLearningSubset,KNN|K-NN,,K-nearest Neighbor Algorithm,FALSE,#N/A,,,"A machine learning that groups objects by a plurality vote of its neighbors, assigning each object to the class most common among its k nearest neighbors.","A machine learning task that groups objects by a plurality vote of its neighbors, assigning each object to the class most common among its k nearest neighbors.",https://en.wikipedia.org/wiki/K-nearest_neighbors_algorithm|,,,AIO:MachineLearningTask +AIO:SelfsupervisedLearning,AIO:MachineLearningSubset,,,Self-supervised Learning,FALSE,#N/A,,,A machine learning that is intermediate between supervised and unsupervised learning and predicts parts of the input data from other observed parts without relying on human-annotated labels.,A machine learning task that is intermediate between supervised and unsupervised learning and predicts parts of the input data from other observed parts without relying on human-annotated labels.,https://en.wikipedia.org/wiki/Self-supervised_learning|,,,AIO:MachineLearningTask +AIO:MultimodalLearning,AIO:NetworkSubset,,,Multimodal Learning,FALSE,0.42,"A type of deep learning that uses multiple modalities of data, such as text, audio, and images, to improve learning outcomes.","A type of deep learning that uses multiple modalities of data, such as text, audio, and images, to improve learning outcomes.",A type of machine learning that uses multiple modalities of data such as text audio and images to improve learning outcomes.,A machine learning task that uses multiple modalities of data such as text audio and images to improve learning outcomes.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:MachineLearningTask +AIO:ExponentialFunction,AIO:FunctionSubset,,,Exponential Function,FALSE,0.47,The exponential function is a mathematical function denoted by f(x)=exp or e^{x}.,The exponential function is a mathematical function denoted by f(x)=exp or e^{x}.,An activation function that is the mathematical function denoted by f(x)=exp or e^{x}.,A mathematical function denoted by f(x)=exp or e^{x}.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/exponential,,,AIO:MathematicalFunction +AIO:GELUFunction,AIO:FunctionSubset,Gaussian Error Linear Unit|GELU,,GELU Function,FALSE,0.58,"Gaussian error linear unit (GELU) computes x * P(X <= x), where P(X) ~ N(0, 1). The (GELU) nonlinearity weights inputs by their value, rather than gates inputs by their sign as in ReLU.","Gaussian error linear unit (GELU) computes x * P(X <= x), where P(X) ~ N(0, 1). The (GELU) nonlinearity weights inputs by their value, rather than gates inputs by their sign as in ReLU.",An activation function that computes x * P(X <= x) where P(X) ~ N(0 1) weighting inputs by their value rather than gating inputs by their sign as in ReLU.,A mathematical function that computes x * P(X <= x) where P(X) ~ N(0 1) weighting inputs by their value rather than gating inputs by their sign as in ReLU.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/gelu,,,AIO:MathematicalFunction +AIO:HardSigmoidFunction,AIO:FunctionSubset,,,Hard Sigmoid Function,FALSE,0.71,A faster approximation of the sigmoid activation. Piecewise linear approximation of the sigmoid function. Ref: 'https://en.wikipedia.org/wiki/Hard_sigmoid',A faster approximation of the sigmoid activation. Piecewise linear approximation of the sigmoid function. Ref: 'https://en.wikipedia.org/wiki/Hard_sigmoid',An activation function that is a faster approximation of the sigmoid activation using a piecewise linear approximation.,A mathematical function that is a faster approximation of the sigmoid activation using a piecewise linear approximation.,https://www.tensorflow.org/api_docs/python/tf/keras/activations/hard_sigmoid,,,AIO:MathematicalFunction +AIO:MathematicalFunction,AIO:ClassSubset,,,Mathematical Function,FALSE,#N/A,A mathematical rule that gives the value of a dependent variable corresponding to specified values of one or more independent variables.,,#N/A,A mathematical rule that gives the value of a dependent variable corresponding to specified values of one or more independent variables.,https://www.sciencedirect.com/topics/mathematics/mathematical-function,,,owl:Thing +AIO:AddLayer,AIO:LayerSubset,,,Add Layer,FALSE,0.66,"Layer that adds a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).","Layer that adds a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).",A merging layer that adds a list of inputs taking as input a list of tensors all of the same shape.,A merging layer that adds a list of inputs taking as input a list of tensors all of the same shape.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Add,,,AIO:MergingLayer +AIO:AverageLayer,AIO:LayerSubset,,,Average Layer,FALSE,0.63,"Layer that averages a list of inputs element-wise. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).","Layer that averages a list of inputs element-wise. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).",A merging layer that averages a list of inputs element-wise taking as input a list of tensors all of the same shape.,A merging layer that averages a list of inputs element-wise taking as input a list of tensors all of the same shape.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Average,,,AIO:MergingLayer +AIO:MaximumLayer,AIO:LayerSubset,,,Maximum Layer,FALSE,0.67,"Layer that computes the maximum (element-wise) a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).","Layer that computes the maximum (element-wise) a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).",A merging layer that computes the maximum (element-wise) of a list of inputs.,A merging layer that computes the maximum (element-wise) of a list of inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Maximum,,,AIO:MergingLayer +AIO:MinimumLayer,AIO:LayerSubset,,,Minimum Layer,FALSE,0.67,"Layer that computes the minimum (element-wise) a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).","Layer that computes the minimum (element-wise) a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).",A merging layer that computes the minimum (element-wise) of a list of inputs.,A merging layer that computes the minimum (element-wise) of a list of inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Minimum,,,AIO:MergingLayer +AIO:ConcatenateLayer,AIO:LayerSubset,,,Concatenate Layer,FALSE,0.56,"Layer that concatenates a list of inputs. It takes as input a list of tensors, all of the same shape except for the concatenation axis, and returns a single tensor that is the concatenation of all inputs.","Layer that concatenates a list of inputs. It takes as input a list of tensors, all of the same shape except for the concatenation axis, and returns a single tensor that is the concatenation of all inputs.",A merging layer that concatenates a list of inputs taking as input a list of tensors all of the same shape except for the concatenation axis.,A merging layer that concatenates a list of inputs taking as input a list of tensors all of the same shape except for the concatenation axis.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Concatenate,,,AIO:MergingLayer +AIO:MultiplyLayer,AIO:LayerSubset,,,Multiply Layer,FALSE,0.73,"Layer that multiplies (element-wise) a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).","Layer that multiplies (element-wise) a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).",A merging layer that multiplies (element-wise) a list of inputs.,A merging layer that multiplies (element-wise) a list of inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Multiply,,,AIO:MergingLayer +AIO:SubtractLayer,AIO:LayerSubset,,,Subtract Layer,FALSE,0.87,"Layer that subtracts two inputs. It takes as input a list of tensors of size 2, both of the same shape, and returns a single tensor, (inputs[0] - inputs[1]), also of the same shape.","Layer that subtracts two inputs. It takes as input a list of tensors of size 2, both of the same shape, and returns a single tensor, (inputs[0] - inputs[1]), also of the same shape.",A merging layer that subtracts two inputs.,A merging layer that subtracts two inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Subtract,,,AIO:MergingLayer +AIO:LanguageModel,AIO:ModelSubset,,,Language Model,FALSE,0.09,A language model is a probabilistic model designed to predict the next word in a sequence or assign probabilities to sequences of words in natural language.,,A model designed to predict the next word in a sequence or assign probabilities to sequences of words in natural language.,A model designed to predict the next word in a sequence or assign probabilities to sequences of words in natural language.,https://en.wikipedia.org/wiki/Language_model,,,AIO:Model +AIO:ThresholdAutoregressive,AIO:ModelSubset,TAR,,Threshold Autoregressive,FALSE,#N/A,,,"A model that allows for different autoregressive processes depending on the regime or state of the time series, enabling the capture of nonlinear behaviors.","A model that allows for different autoregressive processes depending on the regime or state of the time series, enabling the capture of nonlinear behaviors.",https://dx.doi.org/10.1080/01621459.1989.10478760,,,AIO:Model +AIO:DynamicConditionalCorrelation,AIO:ModelSubset,DCC,,Dynamic Conditional Correlation,FALSE,#N/A,,,"A model that allows for time-varying correlations between different time series, used in financial econometrics to model and forecast covariances.","A model that allows for time-varying correlations between different time series, used in financial econometrics to model and forecast covariances.",,,,AIO:Model +AIO:VectorAutoregression,AIO:ModelSubset,VAR,,Vector Autoregression,FALSE,#N/A,,,"A model that captures the linear interdependencies among multiple time series, where each variable is modeled as a linear function of its own past values and the past values of all other variables in the system.","A model that captures the linear interdependencies among multiple time series, where each variable is modeled as a linear function of its own past values and the past values of all other variables in the system.",,,,AIO:Model +AIO:AutoregressiveMovingAverage,AIO:ModelSubset,ARMA,,Autoregressive Moving Average,FALSE,#N/A,,,"A model that combines autoregressive (AR) and moving average (MA) components to represent time series data, suitable for stationary series without the need for differencing.","A model that combines autoregressive (AR) and moving average (MA) components to represent time series data, suitable for stationary series without the need for differencing.",,,,AIO:Model +AIO:ExponentialSmoothingStateSpaceModel,AIO:ModelSubset,ETS,,Exponential Smoothing State Space Model,FALSE,#N/A,,,"A model that combines exponential smoothing with state space modeling, allowing for the inclusion of both trend and seasonal components. Used in forecasting.","A model that combines exponential smoothing with state space modeling, allowing for the inclusion of both trend and seasonal components. Used in forecasting.",,,,AIO:Model +AIO:AutoregressiveConditionalHeteroskedasticity,AIO:ModelSubset,ARCH,,Autoregressive Conditional Heteroskedasticity,FALSE,#N/A,,,"A model that describes the variance of the current error term as a function of the previous periods' error terms, capturing volatility clustering. Used for time series data.","A model that describes the variance of the current error term as a function of the previous periods' error terms, capturing volatility clustering. Used for time series data.",,,,AIO:Model +AIO:SeasonalAutoregressiveIntegratedMovingAverage,AIO:ModelSubset,SARIMA,,Seasonal Autoregressive Integrated Moving-Average,FALSE,#N/A,,,"A model that extends ARIMA, explicitly supporting univariate time series data with a seasonal component, combining seasonal differencing with ARIMA modeling.","A model that extends ARIMA, explicitly supporting univariate time series data with a seasonal component, combining seasonal differencing with ARIMA modeling.",,,,AIO:Model +AIO:AutoregressiveDistributedLag,AIO:ModelSubset,ARDL,,Autoregressive Distributed Lag,FALSE,#N/A,,,"A model that includes lagged values of both the dependent variable and one or more independent variables, capturing dynamic relationships over time. Used in time series analysis.","A model that includes lagged values of both the dependent variable and one or more independent variables, capturing dynamic relationships over time. Used in time series analysis.",,,,AIO:Model +AIO:GeneralizedAutoregressiveConditionalHeteroskedasticity,AIO:ModelSubset,GARCH,,Generalized Autoregressive Conditional Heteroskedasticity,FALSE,#N/A,,,"A model that incorporates lagged conditional variances, allowing for more flexibility in modeling time-varying volatility.","A model that incorporates lagged conditional variances, allowing for more flexibility in modeling time-varying volatility.",,,,AIO:Model +AIO:AutoregressiveIntegratedMovingAverage,AIO:ModelSubset,ARIMA,,Autoregressive Integrated Moving Average,FALSE,#N/A,,,"A model which combines autoregression (AR), differencing (I), and moving average (MA) components. Used for analyzing and forecasting time series data.","A model which combines autoregression (AR), differencing (I), and moving average (MA) components. Used for analyzing and forecasting time series data.",,,,AIO:Model +AIO:ModularLLM,AIO:ModelSubset,Modular Large Language Model,component skills|skill composition,Modular LLM,FALSE,0.20,,,A modular large language model that consists of multiple specialized components or skills that can be dynamically composed and recombined to solve complex tasks mimicking the modular structure of human cognition.,A modular language model that consists of multiple specialized components or skills that can be dynamically composed and recombined to solve complex tasks mimicking the modular structure of human cognition.,https://arxiv.org/abs/2302.11529v2,,,AIO:ModularLanguageModel +AIO:MultimodalLLM,AIO:ModelSubset,Multimodal Large Language Model,cross-modal grounding,Multimodal LLM,FALSE,0.34,,,A multimodal large language model that learns joint representations across different modalities like text vision and audio in an end-to-end fashion for better cross-modal understanding and generation.,A multimodal large language model that learns joint representations across different modalities like text vision and audio in an end-to-end fashion for better cross-modal understanding and generation.,https://arxiv.org/abs/2303.17580,,,AIO:MultimodalLanguageModel +AIO:MultimodalPromptbasedLanguageModel,AIO:ModelSubset,,,Multimodal Prompt-based Language Model,FALSE,#N/A,,,"A multimodal LLM which processes prompts that include multiple modalities, such as both text and images, to generate relevant responses.","A multimodal large language model which processes prompts that include multiple modalities, such as both text and images, to generate relevant responses.",https://arxiv.org/abs/2210.03094,,,AIO:MultimodalLanguageModel +AIO:ArtificialNeuralNetwork,AIO:NetworkSubset,ANN|NN,,Artificial Neural Network,FALSE,0.58,"An artificial neural network (ANN) is based on a collection of connected units or nodes called artificial neurons, modeled after biological neurons, with connections transmitting signals processed by non-linear functions.","An artificial neural network (ANN) is based on a collection of connected units or nodes called artificial neurons, modeled after biological neurons, with connections transmitting signals processed by non-linear functions.",A network based on a collection of connected units called artificial neurons modeled after biological neurons.,A network based on a collection of connected units called artificial neurons modeled after biological neurons.,https://en.wikipedia.org/wiki/Artificial_neural_network,,,AIO:Network +AIO:UnsupervisedPretrainedNetwork,AIO:NetworkSubset,UPN,,Unsupervised Pretrained Network,FALSE,0.54,"Unsupervised pre-training initializes a discriminative neural net from one trained using an unsupervised criterion, aiding in optimization and overfitting issues.","Unsupervised pre-training initializes a discriminative neural net from one trained using an unsupervised criterion, aiding in optimization and overfitting issues.",A network that initializes a discriminative neural net from one trained using an unsupervised criterion.,A network that initializes a discriminative neural net from one trained using an unsupervised criterion.,https://metacademy.org/graphs/concepts/unsupervised_pre_training,,,AIO:Network +AIO:BayesianNetwork,AIO:NetworkSubset,,,Bayesian Network,FALSE,0.30,A probabilistic graphical model representing variables and their conditional dependencies via a directed acyclic graph (DAG).,,A network that is a probabilistic graphical model representing variables and their conditional dependencies via a directed acyclic graph.,A network that is a probabilistic graphical model representing variables and their conditional dependencies via a directed acyclic graph.,https://en.wikipedia.org/wiki/Bayesian_network,,,AIO:Network +AIO:MarkovChain,AIO:ModelSubset,MC|Markov Process|MP,,Markov Chain,FALSE,0.16,A Markov chain is a stochastic model describing a sequence of possible events where the probability of each event depends only on the previous event's state.,,A network that is a stochastic model describing a sequence of possible events where the probability of each event depends only on the previous event's state.,A network that is a stochastic model describing a sequence of possible events where the probability of each event depends only on the previous event's state.,https://en.wikipedia.org/wiki/Markov_chain,Layers: Probalistic Hidden,AIO:ProbabilisticHiddenLayer,AIO:Network +AIO:SymmetricallyConnectedNetwork,AIO:NetworkSubset,SCN,,Symmetrically Connected Network,FALSE,0.59,"Symmetrically connected networks are a type of recurrent neural network where connections between units are symmetrical, meaning they have equal weights in both directions. This structure allows the network to maintain consistent information flow and equilibrium.","Symmetrically connected networks are a type of recurrent neural network where connections between units are symmetrical, meaning they have equal weights in both directions. This structure allows the network to maintain consistent information flow and equilibrium.",A network that is a type of recurrent neural network where connections between units are symmetrical with equal weights in both directions.,A network that is a type of recurrent neural network where connections between units are symmetrical with equal weights in both directions.,https://ieeexplore.ieee.org/document/287176,,,AIO:Network +AIO:LiquidStateMachineNetwork,AIO:NetworkSubset,LSM,,Liquid State Machine Network,FALSE,0.50,"A liquid state machine (LSM) is a type of reservoir computer using a spiking neural network, with recurrently connected nodes turning time-varying input into spatio-temporal activation patterns.",,A network that is a type of reservoir computer turning time-varying input into spatio-temporal activation patterns.,A network that is a type of reservoir computer turning time-varying input into spatio-temporal activation patterns.,https://en.wikipedia.org/wiki/Liquid_state_machine,"Layers: Input, Spiking Hidden, Output",AIO:InputLayer|AIO:SpikingHiddenLayer|AIO:OutputLayer,AIO:Network +AIO:KohonenNetwork,AIO:NetworkSubset,KN|Self-Organizing Map|SOM|Self-Organizing Feature Map|SOFM,,Kohonen Network,FALSE,0.40,"A self-organizing map (SOM) or Kohonen network is an unsupervised machine learning technique producing a low-dimensional representation of high-dimensional data, preserving topological structure.",,A network that is an unsupervised technique producing a low-dimensional representation of high-dimensional data preserving topological structure.,A network that is an unsupervised technique producing a low-dimensional representation of high-dimensional data preserving topological structure.,https://en.wikipedia.org/wiki/Self-organizing_map,"Layers: Input, Hidden",AIO:InputLayer|AIO:HiddenLayer,AIO:Network +AIO:RecurrentNeuralNetwork,AIO:NetworkSubset,RecNN|Recurrent Network|RN,,Recurrent Neural Network,FALSE,0.35,"A recurrent neural network (RNN) has connections forming a directed graph along a temporal sequence, enabling dynamic temporal behavior.",,A deep neural network with connections forming a directed graph along a temporal sequence enabling dynamic behavior.,A network with connections forming a directed graph along a temporal sequence enabling dynamic behavior.,,,,AIO:Network +AIO:SupportVectorMachine,AIO:NetworkSubset,SVM|Supper Vector Network|SVN,,Support Vector Machine,FALSE,0.45,"Support vector machines (SVMs) are supervised learning models for classification and regression analysis, mapping training examples to points in space to maximize the gap between categories.",,A network with supervised learning models for classification and regression that maps training examples to points in space maximizing the gap between categories.,A network with supervised learning models for classification and regression that maps training examples to points in space maximizing the gap between categories.,https://en.wikipedia.org/wiki/Support-vector_machine,"Layers: Input, Hidden, Output",AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:Network +AIO:node2vecCBOW,AIO:NetworkSubset,N2V-CBOW,CBOW,node2vec-CBOW,FALSE,#N/A,,,"A node2vec that predicts the current node from a window of surrounding context nodes, with the order of context nodes not influencing prediction.","A node2vec that predicts the current node from a window of surrounding context nodes, with the order of context nodes not influencing prediction.",https://en.wikipedia.org/wiki/Node2vec,"Layers: Input, Hidden, Output",AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:node2vec +AIO:node2vecSkipGram,AIO:NetworkSubset,N2V-SkipGram,SkipGram,node2vec-SkipGram,FALSE,#N/A,,,"A node2vec that uses the current node to predict the surrounding window of context nodes, weighing nearby context nodes more heavily than distant ones.","A node2vec that uses the current node to predict the surrounding window of context nodes, weighing nearby context nodes more heavily than distant ones.",https://en.wikipedia.org/wiki/Node2vec,"Layers: Input, Hidden, Output",AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:node2vec +AIO:GroupNormLayer,AIO:LayerSubset,GroupNorm|GroupNorm,,GroupNorm Layer,FALSE,0.68,Applies Group Normalization over a mini-batch of inputs as described in the paper Group Normalization,Applies Group Normalization over a mini-batch of inputs as described in the paper Group Normalization,A normalization layer that applies Group Normalization over a mini-batch of inputs.,A normalization layer that applies Group Normalization over a mini-batch of inputs.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:NormalizationLayer +AIO:InstanceNorm1DLayer,AIO:LayerSubset,InstanceNorm1D|InstanceNorm1D|InstanceNorm1D,,InstanceNorm1D Layer,FALSE,0.69,Applies Instance Normalization over a 2D (unbatched) or 3D (batched) input as described in the paper Instance Normalization: The Missing Ingredient for Fast Stylization.,Applies Instance Normalization over a 2D (unbatched) or 3D (batched) input as described in the paper Instance Normalization: The Missing Ingredient for Fast Stylization.,A normalization layer that applies Instance Normalization over a 2D (unbatched) or 3D (batched) input.,A normalization layer that applies Instance Normalization over a 2D (unbatched) or 3D (batched) input.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:NormalizationLayer +AIO:InstanceNorm2D,AIO:LayerSubset,,,InstanceNorm2D,FALSE,0.59,Applies Instance Normalization over a 4D input (a mini-batch of 2D inputs with additional channel dimension) as described in the paper Instance Normalization: The Missing Ingredient for Fast Stylization.,Applies Instance Normalization over a 4D input (a mini-batch of 2D inputs with additional channel dimension) as described in the paper Instance Normalization: The Missing Ingredient for Fast Stylization.,A normalization layer that applies Instance Normalization over a 4D input (a mini-batch of 2D inputs with additional channel dimension).,A normalization layer that applies Instance Normalization over a 4D input (a mini-batch of 2D inputs with additional channel dimension).,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:NormalizationLayer +AIO:InstanceNorm3DLayer,AIO:LayerSubset,InstanceNorm3D,,InstanceNorm3D Layer,FALSE,0.59,Applies Instance Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension) as described in the paper Instance Normalization: The Missing Ingredient for Fast Stylization.,Applies Instance Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension) as described in the paper Instance Normalization: The Missing Ingredient for Fast Stylization.,A normalization layer that applies Instance Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension).,A normalization layer that applies Instance Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension).,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:NormalizationLayer +AIO:LayerNormLayer,AIO:LayerSubset,LayerNorm,,LayerNorm Layer,FALSE,0.68,Applies Layer Normalization over a mini-batch of inputs as described in the paper Layer Normalization,Applies Layer Normalization over a mini-batch of inputs as described in the paper Layer Normalization,A normalization layer that applies Layer Normalization over a mini-batch of inputs.,A normalization layer that applies Layer Normalization over a mini-batch of inputs.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:NormalizationLayer +AIO:LayerNormalizationLayer,AIO:LayerSubset,,,LayerNormalization Layer,FALSE,0.89,"Layer normalization layer (Ba et al., 2016). Normalize the activations of the previous layer for each given example in a batch independently, rather than across a batch like Batch Normalization. i.e. applies a transformation that maintains the mean activation within each example close to 0 and the activation standard deviation close to 1. Given a tensor inputs, moments are calculated and normalization is performed across the axes specified in axis.","Layer normalization layer (Ba et al., 2016). Normalize the activations of the previous layer for each given example in a batch independently, rather than across a batch like Batch Normalization. i.e. applies a transformation that maintains the mean activation within each example close to 0 and the activation standard deviation close to 1. Given a tensor inputs, moments are calculated and normalization is performed across the axes specified in axis.",A normalization layer that applies Layer Normalization over the inputs.,A normalization layer that applies Layer Normalization over the inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/LayerNormalization,,,AIO:NormalizationLayer +AIO:LocalResponseNormLayer,AIO:LayerSubset,LocalResponseNorm,,LocalResponseNorm Layer,FALSE,0.57,"Applies local response normalization over an input signal composed of several input planes, where channels occupy the second dimension.","Applies local response normalization over an input signal composed of several input planes, where channels occupy the second dimension.",A normalization layer that applies local response normalization over an input signal composed of several input planes.,A normalization layer that applies local response normalization over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#normalization-layers,,,AIO:NormalizationLayer +AIO:BatchNormalizationLayer,AIO:LayerSubset,BatchNorm,,BatchNormalization Layer,FALSE,0.87,"Layer that normalizes its inputs. Batch normalization applies a transformation that maintains the mean output close to 0 and the output standard deviation close to 1. Importantly, batch normalization works differently during training and during inference. During training (i.e. when using fit() or when calling the layer/model with the argument training=True), the layer normalizes its output using the mean and standard deviation of the current batch of inputs. That is to say, for each channel being normalized, the layer returns gamma * (batch - mean(batch)) / sqrt(var(batch) + epsilon) + beta, where: epsilon is small constant (configurable as part of the constructor arguments), gamma is a learned scaling factor (initialized as 1), which can be disabled by passing scale=False to the constructor. beta is a learned offset factor (initialized as 0), which can be disabled by passing center=False to the constructor. During inference (i.e. when using evaluate() or predict() or when calling the layer/model with the argument training=False (which is the default), the layer normalizes its output using a moving average of the mean and standard deviation of the batches it has seen during training. That is to say, it returns gamma * (batch - self.moving_mean) / sqrt(self.moving_var + epsilon) + beta. self.moving_mean and self.moving_var are non-trainable variables that are updated each time the layer in called in training mode, as such: moving_mean = moving_mean * momentum + mean(batch) * (1 - momentum) moving_var = moving_var * momentum + var(batch) * (1 - momentum).","Layer that normalizes its inputs. Batch normalization applies a transformation that maintains the mean output close to 0 and the output standard deviation close to 1. Importantly, batch normalization works differently during training and during inference. During training (i.e. when using fit() or when calling the layer/model with the argument training=True), the layer normalizes its output using the mean and standard deviation of the current batch of inputs. That is to say, for each channel being normalized, the layer returns gamma * (batch - mean(batch)) / sqrt(var(batch) + epsilon) + beta, where: epsilon is small constant (configurable as part of the constructor arguments), gamma is a learned scaling factor (initialized as 1), which can be disabled by passing scale=False to the constructor. beta is a learned offset factor (initialized as 0), which can be disabled by passing center=False to the constructor. During inference (i.e. when using evaluate() or predict() or when calling the layer/model with the argument training=False (which is the default), the layer normalizes its output using a moving average of the mean and standard deviation of the batches it has seen during training. That is to say, it returns gamma * (batch - self.moving_mean) / sqrt(self.moving_var + epsilon) + beta. self.moving_mean and self.moving_var are non-trainable variables that are updated each time the layer in called in training mode, as such: moving_mean = moving_mean * momentum + mean(batch) * (1 - momentum) moving_var = moving_var * momentum + var(batch) * (1 - momentum).",A normalization layer that normalizes its inputs applying a transformation that maintains the mean close to 0 and the standard deviation close to 1.,A normalization layer that normalizes its inputs applying a transformation that maintains the mean close to 0 and the standard deviation close to 1.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/BatchNormalization,,,AIO:NormalizationLayer +AIO:NormalizationLayer,AIO:LayerSubset,,,Normalization Layer,FALSE,0.25,A preprocessing layer which normalizes continuous features.,,A preprocessing layer that normalizes continuous features.,A numerical features prepreprocessing layer that normalizes continuous features.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Normalization,,,AIO:NumericalFeaturesPreprocessingLayer +AIO:DiscretizationLayer,AIO:LayerSubset,,,Discretization Layer,FALSE,0.00,A preprocessing layer which buckets continuous features by ranges.,,A preprocessing layer which buckets continuous features by ranges.,A numerical features prepreprocessing layer which buckets continuous features by ranges.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Discretization,,,AIO:NumericalFeaturesPreprocessingLayer +AIO:LPPool1DLayer,AIO:LayerSubset,LPPool1D,,LPPool1D Layer,FALSE,0.35,Applies a 1D power-average pooling over an input signal composed of several input planes.,,A pooling layer that applies 1D power-average pooling over an input signal composed of several input planes.,A pooling layer that applies 1D power-average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:LPPool2DLayer,AIO:LayerSubset,LPPool2D,,LPPool2D Layer,FALSE,0.35,Applies a 2D power-average pooling over an input signal composed of several input planes.,,A pooling layer that applies 2D power-average pooling over an input signal composed of several input planes.,A pooling layer that applies 2D power-average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:AdaptiveAvgPool1DLayer,AIO:LayerSubset,AdaptiveAvgPool1D,,AdaptiveAvgPool1D Layer,FALSE,0.28,Applies a 1D adaptive average pooling over an input signal composed of several input planes.,,A pooling layer that applies a 1D adaptive average pooling over an input signal composed of several input planes.,A pooling layer that applies a 1D adaptive average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:AdaptiveMaxPool1DLayer,AIO:LayerSubset,AdaptiveMaxPool1D,,AdaptiveMaxPool1D Layer,FALSE,0.28,Applies a 1D adaptive max pooling over an input signal composed of several input planes.,,A pooling layer that applies a 1D adaptive max pooling over an input signal composed of several input planes.,A pooling layer that applies a 1D adaptive max pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:AvgPool1DLayer,AIO:LayerSubset,AvgPool1D,,AvgPool1D Layer,FALSE,0.29,Applies a 1D average pooling over an input signal composed of several input planes.,,A pooling layer that applies a 1D average pooling over an input signal composed of several input planes.,A pooling layer that applies a 1D average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:AdaptiveAvgPool2DLayer,AIO:LayerSubset,AdaptiveAvgPool2D,,AdaptiveAvgPool2D Layer,FALSE,0.28,Applies a 2D adaptive average pooling over an input signal composed of several input planes.,,A pooling layer that applies a 2D adaptive average pooling over an input signal composed of several input planes.,A pooling layer that applies a 2D adaptive average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:AdaptiveMaxPool2DLayer,AIO:LayerSubset,AdaptiveMaxPool2D,,AdaptiveMaxPool2D Layer,FALSE,0.28,Applies a 2D adaptive max pooling over an input signal composed of several input planes.,,A pooling layer that applies a 2D adaptive max pooling over an input signal composed of several input planes.,A pooling layer that applies a 2D adaptive max pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:AvgPool2DLayer,AIO:LayerSubset,AvgPool2D,,AvgPool2D Layer,FALSE,0.29,Applies a 2D average pooling over an input signal composed of several input planes.,,A pooling layer that applies a 2D average pooling over an input signal composed of several input planes.,A pooling layer that applies a 2D average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:FractionalMaxPool2DLayer,AIO:LayerSubset,FractionalMaxPool2D,,FractionalMaxPool2D Layer,FALSE,0.28,Applies a 2D fractional max pooling over an input signal composed of several input planes.,,A pooling layer that applies a 2D fractional max pooling over an input signal composed of several input planes.,A pooling layer that applies a 2D fractional max pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:AdaptiveAvgPool3DLayer,AIO:LayerSubset,AdaptiveAvgPool3D,,AdaptiveAvgPool3D Layer,FALSE,0.28,Applies a 3D adaptive average pooling over an input signal composed of several input planes.,,A pooling layer that applies a 3D adaptive average pooling over an input signal composed of several input planes.,A pooling layer that applies a 3D adaptive average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:AdaptiveMaxPool3DLayer,AIO:LayerSubset,AdaptiveMaxPool3D,,AdaptiveMaxPool3D Layer,FALSE,0.28,Applies a 3D adaptive max pooling over an input signal composed of several input planes.,,A pooling layer that applies a 3D adaptive max pooling over an input signal composed of several input planes.,A pooling layer that applies a 3D adaptive max pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:AvgPool3DLayer,AIO:LayerSubset,AvgPool3D,,AvgPool3D Layer,FALSE,0.29,Applies a 3D average pooling over an input signal composed of several input planes.,,A pooling layer that applies a 3D average pooling over an input signal composed of several input planes.,A pooling layer that applies a 3D average pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:FractionalMaxPool3DLayer,AIO:LayerSubset,FractionalMaxPool3D,,FractionalMaxPool3D Layer,FALSE,0.28,Applies a 3D fractional max pooling over an input signal composed of several input planes.,,A pooling layer that applies a 3D fractional max pooling over an input signal composed of several input planes.,A pooling layer that applies a 3D fractional max pooling over an input signal composed of several input planes.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:MaxUnpool1DLayer,AIO:LayerSubset,MaxUnpool1D,,MaxUnpool1D Layer,FALSE,0.55,Computes a partial inverse of MaxPool1D.,Computes a partial inverse of MaxPool1D.,A pooling layer that computes a partial inverse of MaxPool1D.,A pooling layer that computes a partial inverse of MaxPool1D.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:MaxUnpool2DLayer,AIO:LayerSubset,MaxUnpool2D,,MaxUnpool2D Layer,FALSE,0.55,Computes a partial inverse of MaxPool2D.,Computes a partial inverse of MaxPool2D.,A pooling layer that computes a partial inverse of MaxPool2D.,A pooling layer that computes a partial inverse of MaxPool2D.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:MaxUnpool3DLayer,AIO:LayerSubset,MaxUnpool3D,,MaxUnpool3D Layer,FALSE,0.55,Computes a partial inverse of MaxPool3D.,Computes a partial inverse of MaxPool3D.,A pooling layer that computes a partial inverse of MaxPool3D.,A pooling layer that computes a partial inverse of MaxPool3D.,https://pytorch.org/docs/stable/nn.html#pooling-layers,,,AIO:PoolingLayer +AIO:AveragePooling3DLayer,AIO:LayerSubset,AvgPool3D,,AveragePooling3D Layer,FALSE,0.82,"Average pooling operation for 3D data (spatial or spatio-temporal). Downsamples the input along its spatial dimensions (depth, height, and width) by taking the average value over an input window (of size defined by pool_size) for each channel of the input. The window is shifted by strides along each dimension.","Average pooling operation for 3D data (spatial or spatio-temporal). Downsamples the input along its spatial dimensions (depth, height, and width) by taking the average value over an input window (of size defined by pool_size) for each channel of the input. The window is shifted by strides along each dimension.",A pooling layer that performs average pooling for 3D data (spatial or spatio-temporal).,A pooling layer that performs average pooling for 3D data (spatial or spatio-temporal).,https://www.tensorflow.org/api_docs/python/tf/keras/layers/AveragePooling3D,,,AIO:PoolingLayer +AIO:AveragePooling2DLayer,AIO:LayerSubset,AvgPool2D,,AveragePooling2D Layer,FALSE,0.93,"Average pooling operation for spatial data. Downsamples the input along its spatial dimensions (height and width) by taking the average value over an input window (of size defined by pool_size) for each channel of the input. The window is shifted by strides along each dimension. The resulting output when using ""valid"" padding option has a shape (number of rows or columns) of: output_shape = math.floor((input_shape - pool_size) / strides) + 1 (when input_shape >= pool_size). The resulting output shape when using the ""same"" padding option is: output_shape = math.floor((input_shape - 1) / strides) + 1.","Average pooling operation for spatial data. Downsamples the input along its spatial dimensions (height and width) by taking the average value over an input window (of size defined by pool_size) for each channel of the input. The window is shifted by strides along each dimension. The resulting output when using ""valid"" padding option has a shape (number of rows or columns) of: output_shape = math.floor((input_shape - pool_size) / strides) + 1 (when input_shape >= pool_size). The resulting output shape when using the ""same"" padding option is: output_shape = math.floor((input_shape - 1) / strides) + 1.",A pooling layer that performs average pooling for spatial data.,A pooling layer that performs average pooling for spatial data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/AveragePooling2D,,,AIO:PoolingLayer +AIO:AveragePooling1DLayer,AIO:LayerSubset,AvgPool1D,,AveragePooling1D Layer,FALSE,0.90,"Average pooling for temporal data. Downsamples the input representation by taking the average value over the window defined by pool_size. The window is shifted by strides. The resulting output when using ""valid"" padding option has a shape of: output_shape = (input_shape - pool_size + 1) / strides). The resulting output shape when using the ""same"" padding option is: output_shape = input_shape / strides.","Average pooling for temporal data. Downsamples the input representation by taking the average value over the window defined by pool_size. The window is shifted by strides. The resulting output when using ""valid"" padding option has a shape of: output_shape = (input_shape - pool_size + 1) / strides). The resulting output shape when using the ""same"" padding option is: output_shape = input_shape / strides.",A pooling layer that performs average pooling for temporal data.,A pooling layer that performs average pooling for temporal data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/AveragePooling1D,,,AIO:PoolingLayer +AIO:GlobalAveragePooling3DLayer,AIO:LayerSubset,GlobalAvgPool3D,,GlobalAveragePooling3D Layer,FALSE,0.62,Global Average pooling operation for 3D data.,Global Average pooling operation for 3D data.,A pooling layer that performs global average pooling operation for 3D data.,A pooling layer that performs global average pooling operation for 3D data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalAveragePooling3D,,,AIO:PoolingLayer +AIO:GlobalAveragePooling2DLayer,AIO:LayerSubset,GlobalAvgPool2D,,GlobalAveragePooling2D Layer,FALSE,0.50,Global average pooling operation for spatial data.,Global average pooling operation for spatial data.,A pooling layer that performs global average pooling operation for spatial data.,A pooling layer that performs global average pooling operation for spatial data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalAveragePooling2D,,,AIO:PoolingLayer +AIO:GlobalAveragePooling1DLayer,AIO:LayerSubset,GlobalAvgPool1D,,GlobalAveragePooling1D Layer,FALSE,0.50,Global average pooling operation for temporal data.,Global average pooling operation for temporal data.,A pooling layer that performs global average pooling operation for temporal data.,A pooling layer that performs global average pooling operation for temporal data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalAveragePooling1D,,,AIO:PoolingLayer +AIO:GlobalMaxPooling3DLayer,AIO:LayerSubset,GlobalMaxPool3D,,GlobalMaxPooling3D Layer,FALSE,0.62,Global Max pooling operation for 3D data.,Global Max pooling operation for 3D data.,A pooling layer that performs global max pooling operation for 3D data.,A pooling layer that performs global max pooling operation for 3D data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalMaxPool3D,,,AIO:PoolingLayer +AIO:GlobalMaxPooling2DLayer,AIO:LayerSubset,GlobalMaxPool2D,,GlobalMaxPooling2D Layer,FALSE,0.50,Global max pooling operation for spatial data.,Global max pooling operation for spatial data.,A pooling layer that performs global max pooling operation for spatial data.,A pooling layer that performs global max pooling operation for spatial data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalMaxPool2D,,,AIO:PoolingLayer +AIO:GlobalMaxPooling1DLayer,AIO:LayerSubset,GlobalMaxPool1D,,GlobalMaxPooling1D Layer,FALSE,0.54,Global max pooling operation for 1D temporal data.,Global max pooling operation for 1D temporal data.,A pooling layer that performs global max pooling operation for temporal data.,A pooling layer that performs global max pooling operation for temporal data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalMaxPool1D,,,AIO:PoolingLayer +AIO:MaxPooling3DLayer,AIO:LayerSubset,MaxPool3D|MaxPooling3D,,MaxPooling3D Layer,FALSE,0.82,"Max pooling operation for 3D data (spatial or spatio-temporal). Downsamples the input along its spatial dimensions (depth, height, and width) by taking the maximum value over an input window (of size defined by pool_size) for each channel of the input. The window is shifted by strides along each dimension.","Max pooling operation for 3D data (spatial or spatio-temporal). Downsamples the input along its spatial dimensions (depth, height, and width) by taking the maximum value over an input window (of size defined by pool_size) for each channel of the input. The window is shifted by strides along each dimension.",A pooling layer that performs max pooling operation for 3D data (spatial or spatio-temporal).,A pooling layer that performs max pooling operation for 3D data (spatial or spatio-temporal).,https://www.tensorflow.org/api_docs/python/tf/keras/layers/MaxPool3D,,,AIO:PoolingLayer +AIO:MaxPooling2DLayer,AIO:LayerSubset,MaxPool2D|MaxPooling2D,,MaxPooling2D Layer,FALSE,0.58,Max pooling operation for 2D spatial data.,Max pooling operation for 2D spatial data.,A pooling layer that performs max pooling operation for spatial data.,A pooling layer that performs max pooling operation for spatial data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/MaxPool2D,,,AIO:PoolingLayer +AIO:MaxPooling1DLayer,AIO:LayerSubset,MaxPool1D|MaxPooling1D,,MaxPooling1D Layer,FALSE,0.91,"Max pooling operation for 1D temporal data. Downsamples the input representation by taking the maximum value over a spatial window of size pool_size. The window is shifted by strides. The resulting output, when using the ""valid"" padding option, has a shape of: output_shape = (input_shape - pool_size + 1) / strides) The resulting output shape when using the ""same"" padding option is: output_shape = input_shape / strides.","Max pooling operation for 1D temporal data. Downsamples the input representation by taking the maximum value over a spatial window of size pool_size. The window is shifted by strides. The resulting output, when using the ""valid"" padding option, has a shape of: output_shape = (input_shape - pool_size + 1) / strides) The resulting output shape when using the ""same"" padding option is: output_shape = input_shape / strides.",A pooling layer that performs max pooling operation for temporal data.,A pooling layer that performs max pooling operation for temporal data.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/MaxPool1D,,,AIO:PoolingLayer +AIO:ModelEfficiency,AIO:PreprocessingSubset,Model Optimization|Computational Efficiency,,Model Efficiency,FALSE,0.17,"Techniques aimed at making models more efficient, such as knowledge distillation.",,Techniques aimed at making models more efficient such as knowledge distillation.,"A preprocessing aimed at making models more efficient, such as knowledge distillation.",https://doi.org/10.1145/3578938|,,,AIO:Preprocessing +AIO:DataPreparation,AIO:PreprocessingSubset,Data Curation|Data Processing|Data Assembly,,Data Preparation,FALSE,0.23,"The process of cleaning, transforming, and organizing raw data into a suitable format for analysis and modeling, ensuring the quality and relevance of the data for machine learning tasks.",,The process of cleaning transforming and organizing raw data into a suitable format for analysis and modeling ensuring the quality and relevance of the data for machine learning tasks.,"A preprocessing that cleans, transforms and organizes raw data into a suitable format for analysis and modeling, ensuring the quality and relevance of the data for machine learning tasks.",,,,AIO:Preprocessing +AIO:Distillation,AIO:PreprocessingSubset,Refining|Purification,Knowledge compression|Teacher-student model,Distillation,FALSE,0.04,"The process of training a smaller model to replicate the behavior of a larger model, aiming to compress the knowledge into a more compact form without significant loss of performance.",,The process of training a smaller model to replicate the behavior of a larger model aiming to compress the knowledge into a more compact form without significant loss of performance.,A preprocessing that trains a smaller model to replicate the behavior of a larger model aiming to compress the knowledge into a more compact form without significant loss of performance.,https://doi.org/10.48550/arXiv.2105.13093|,,,AIO:Preprocessing +AIO:DataEnhancement,AIO:PreprocessingSubset,,,DataEnhancement,FALSE,0.38,"Techniques used to improve the quality, diversity, and volume of data available for training machine learning models, such as data augmentation, synthesis, and enrichment, to enhance model robustness and accuracy.",,Techniques used to improve the quality diversity and volume of data available for training machine learning models such as data augmentation synthesis and enrichment to enhance model robustness and accuracy.,"A preprocessing used to improve the quality diversity and volume of data available for training machine learning models, such as data augmentation synthesis and enrichment to enhance model robustness and accuracy.",,,,AIO:Preprocessing +AIO:TrainingStrategies,AIO:PreprocessingSubset,Learning Techniques|Instructional Methods,,Training Strategies,FALSE,0.12,"The methodologies and approaches used to train machine learning models, including techniques such as supervised learning, unsupervised learning, reinforcement learning, and transfer learning, aimed at optimizing model performance.",,The methodologies and approaches used to train machine learning models including techniques such as supervised learning unsupervised learning reinforcement learning and transfer learning aimed at optimizing model performance.,A preprocessing used to train machine learning models including techniques such as supervised learning unsupervised learning reinforcement learning and transfer learning aimed at optimizing model performance.,,,,AIO:Preprocessing +AIO:ProbabilisticTopicModel,AIO:MachineLearningSubset,,,Probabilistic Topic Model,FALSE,0.39,"Methods that use statistical techniques to analyze the words in each text to discover common themes, their connections, and their changes over time.",,A probabilistic graphical model that uses statistical techniques to analyze the words in each text to discover common themes their connections and their changes over time.,A probabilistic graphical model that uses statistical techniques to analyze the words in each text to discover common themes their connections and their changes over time.,https://pyro.ai/examples/prodlda.html|,,,AIO:ProbabilisticGraphicalModel +AIO:CausalGraphicalModel,AIO:MachineLearningSubset,Casaul Graph|Path Diagram|Casaul Bayesian Network|DAG|Directed Acyclic Graph,,Causal Graphical Model,FALSE,0.36,Probabilistic graphical models used to encode assumptions about the data-generating process.,,A probabilistic graphical model used to encode assumptions about the data-generating process.,A probabilistic graphical model used to encode assumptions about the data-generating process.,https://en.wikipedia.org/wiki/Causal_graph|,,,AIO:ProbabilisticGraphicalModel +AIO:InheritedBias,AIO:BiasSubset,,,Inherited Bias,FALSE,0.30,"Bias arising when machine learning applications generate inputs for other machine learning algorithms, passing on any existing bias.",,A processing bias arising when machine learning applications generate inputs for other machine learning algorithms passing on any existing bias.,A processing bias arising when machine learning applications generate inputs for other machine learning algorithms passing on any existing bias.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:ProcessingBias +AIO:AmplificationBias,AIO:BiasSubset,,,Amplification Bias,FALSE,0.22,Bias arising when the distribution over prediction outputs is skewed compared to the prior distribution of the prediction target.,,A processing bias arising when the distribution over prediction outputs is skewed compared to the prior distribution of the prediction target.,A processing bias arising when the distribution over prediction outputs is skewed compared to the prior distribution of the prediction target.,https://royalsocietypublishing.org/doi/10.1098/rspb.2019.0165#d1e5237,,,AIO:ProcessingBias +AIO:ErrorPropagationBias,AIO:BiasSubset,Error Propagation,,Error Propagation Bias,FALSE,0.33,"The effect of variables' uncertainties (or errors, more specifically random errors) on the uncertainty of a function based on them.",,A processing bias characterized by the effect of variables' uncertainties (or errors more specifically random errors) on the uncertainty of a function based on them.,A processing bias characterized by the effect of variables' uncertainties (or errors more specifically random errors) on the uncertainty of a function based on them.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:ProcessingBias +AIO:SurvivorshipBias,AIO:BiasSubset,,,Survivorship Bias,FALSE,0.48,"The tendency to focus on items, observations, or people that ""survive"" a selection process, overlooking those that did not.","The tendency to focus on items, observations, or people that ""survive"" a selection process, overlooking those that did not.","A processing bias characterized by the tendency to focus on items observations or people that ""survive"" a selection process overlooking those that did not.","A processing bias characterized by the tendency to focus on items observations or people that ""survive"" a selection process overlooking those that did not.",https://doi.org/10.6028/NIST.SP.1270,,,AIO:ProcessingBias +AIO:ModelSelectionBias,AIO:BiasSubset,,,Model Selection Bias,FALSE,0.21,"Bias introduced when using data to select a single ""best"" model from many, or when an explanatory variable has a weak relationship with the response variable.",,"A processing bias introduced when using data to select a single ""best"" model from many or when an explanatory variable has a weak relationship with the response variable.","A processing bias introduced when using data to select a single ""best"" model from many or when an explanatory variable has a weak relationship with the response variable.",https://doi.org/10.6028/NIST.SP.1270,,,AIO:ProcessingBias +AIO:TimeDistributedLayer,AIO:LayerSubset,,,TimeDistributed Layer,FALSE,0.85,"This wrapper allows to apply a layer to every temporal slice of an input. Every input should be at least 3D, and the dimension of index one of the first input will be considered to be the temporal dimension. Consider a batch of 32 video samples, where each sample is a 128x128 RGB image with channels_last data format, across 10 timesteps. The batch input shape is (32, 10, 128, 128, 3). You can then use TimeDistributed to apply the same Conv2D layer to each of the 10 timesteps, independently:","This wrapper allows to apply a layer to every temporal slice of an input. Every input should be at least 3D, and the dimension of index one of the first input will be considered to be the temporal dimension. Consider a batch of 32 video samples, where each sample is a 128x128 RGB image with channels_last data format, across 10 timesteps. The batch input shape is (32, 10, 128, 128, 3). You can then use TimeDistributed to apply the same Conv2D layer to each of the 10 timesteps, independently:",A wrapper layer that applies a layer to every temporal slice of an input.,A recurrent layer that applies a layer to every temporal slice of an input.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/TimeDistributed,,,AIO:RecurrentLayer +AIO:SimpleRNNLayer,AIO:LayerSubset,,,SimpleRNN Layer,FALSE,0.44,Fully-connected RNN where the output is to be fed back to input.,Fully-connected RNN where the output is to be fed back to input.,A recurrent layer that implements a fully-connected RNN where the output is to be fed back to input.,A recurrent layer that implements a fully-connected RNN where the output is to be fed back to input.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/SimpleRNN,,,AIO:RecurrentLayer +AIO:GRULayer,AIO:LayerSubset,,,GRU Layer,FALSE,0.96,"Gated Recurrent Unit - Cho et al. 2014. Based on available runtime hardware and constraints, this layer will choose different implementations (cuDNN-based or pure-TensorFlow) to maximize the performance. If a GPU is available and all the arguments to the layer meet the requirement of the cuDNN kernel (see below for details), the layer will use a fast cuDNN implementation. The requirements to use the cuDNN implementation are: activation == tanh, recurrent_activation == sigmoid, recurrent_dropout == 0, unroll is False, use_bias is True, reset_after is True. Inputs, if use masking, are strictly right-padded. Eager execution is enabled in the outermost context. There are two variants of the GRU implementation. The default one is based on v3 and has reset gate applied to hidden state before matrix multiplication. The other one is based on original and has the order reversed. The second variant is compatible with CuDNNGRU (GPU-only) and allows inference on CPU. Thus it has separate biases for kernel and recurrent_kernel. To use this variant, set reset_after=True and recurrent_activation='sigmoid'.","Gated Recurrent Unit - Cho et al. 2014. Based on available runtime hardware and constraints, this layer will choose different implementations (cuDNN-based or pure-TensorFlow) to maximize the performance. If a GPU is available and all the arguments to the layer meet the requirement of the cuDNN kernel (see below for details), the layer will use a fast cuDNN implementation. The requirements to use the cuDNN implementation are: activation == tanh, recurrent_activation == sigmoid, recurrent_dropout == 0, unroll is False, use_bias is True, reset_after is True. Inputs, if use masking, are strictly right-padded. Eager execution is enabled in the outermost context. There are two variants of the GRU implementation. The default one is based on v3 and has reset gate applied to hidden state before matrix multiplication. The other one is based on original and has the order reversed. The second variant is compatible with CuDNNGRU (GPU-only) and allows inference on CPU. Thus it has separate biases for kernel and recurrent_kernel. To use this variant, set reset_after=True and recurrent_activation='sigmoid'.",A recurrent layer that implements the Gated Recurrent Unit architecture.,A recurrent layer that implements the Gated Recurrent Unit architecture.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GRU,,,AIO:RecurrentLayer +AIO:LSTMLayer,AIO:LayerSubset,,,LSTM Layer,FALSE,0.94,"Long Short-Term Memory layer - Hochreiter 1997. Based on available runtime hardware and constraints, this layer will choose different implementations (cuDNN-based or pure-TensorFlow) to maximize the performance. If a GPU is available and all the arguments to the layer meet the requirement of the cuDNN kernel (see below for details), the layer will use a fast cuDNN implementation. The requirements to use the cuDNN implementation are: 1. activation == tanh, 2. recurrent_activation == sigmoid, 3. recurrent_dropout == 0, 4. unroll is False, 5. use_bias is True, 6. Inputs, if use masking, are strictly right-padded, 7. Eager execution is enabled in the outermost context.","Long Short-Term Memory layer - Hochreiter 1997. Based on available runtime hardware and constraints, this layer will choose different implementations (cuDNN-based or pure-TensorFlow) to maximize the performance. If a GPU is available and all the arguments to the layer meet the requirement of the cuDNN kernel (see below for details), the layer will use a fast cuDNN implementation. The requirements to use the cuDNN implementation are: 1. activation == tanh, 2. recurrent_activation == sigmoid, 3. recurrent_dropout == 0, 4. unroll is False, 5. use_bias is True, 6. Inputs, if use masking, are strictly right-padded, 7. Eager execution is enabled in the outermost context.",A recurrent layer that implements the Long Short-Term Memory architecture.,A recurrent layer that implements the Long Short-Term Memory architecture.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/LSTM,,,AIO:RecurrentLayer +AIO:BidirectionalLayer,AIO:LayerSubset,,,Bidirectional Layer,FALSE,0.73,Bidirectional wrapper for RNNs.,Bidirectional wrapper for RNNs.,A recurrent layer that is a bidirectional wrapper for RNNs.,A recurrent layer that is a bidirectional wrapper for RNNs.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Bidirectional,,,AIO:RecurrentLayer +AIO:UnitNormalizationLayer,AIO:LayerSubset,,,UnitNormalization Layer,FALSE,0.43,Unit normalization layer. Normalize a batch of inputs so that each input in the batch has a L2 norm equal to 1 (across the axes specified in axis).,Unit normalization layer. Normalize a batch of inputs so that each input in the batch has a L2 norm equal to 1 (across the axes specified in axis).,A normalization layer that normalizes a batch of inputs so that each input in the batch has a L2 norm equal to 1.,A recurrent layer that normalizes a batch of inputs so that each input in the batch has a L2 norm equal to 1.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/UnitNormalization,,,AIO:RecurrentLayer +AIO:EchoStateNetwork,AIO:NetworkSubset,ESN,,Echo State Network,FALSE,0.53,"An echo state network (ESN) is a type of reservoir computer with a recurrent neural network and a sparsely connected hidden layer, learning output neuron weights to produce temporal patterns.",,A recurrent neural network with a recurrent hidden layer and sparsely connected hidden neurons that learns output weights to produce temporal patterns.,A recurrent neural network with a recurrent hidden layer and sparsely connected hidden neurons that learns output weights to produce temporal patterns.,https://en.wikipedia.org/wiki/Echo_state_network,"Layers: Input, Recurrent, Output",AIO:InputLayer|AIO:RecurrentLayer|AIO:OutputLayer,AIO:RecurrentNeuralNetwork +AIO:LongShortTermMemory,AIO:NetworkSubset,LSTM,,Long Short Term Memory,FALSE,0.77,"Long short-term memory (LSTM) networks are artificial recurrent neural networks with feedback connections, processing entire sequences of data for tasks like handwriting and speech recognition.",,A recurrent neural network with feedback connections that processes entire sequences of data.,A recurrent neural network with feedback connections that processes entire sequences of data.,https://en.wikipedia.org/wiki/Long_short-term_memory,"Layers: Input, Memory Cell, Output",AIO:InputLayer|AIO:MemoryCellLayer|AIO:OutputLayer,AIO:RecurrentNeuralNetwork +AIO:ProportionalHazardsModel,AIO:MachineLearningSubset,,,Proportional Hazards Model,FALSE,0.17,A survival modeling method where the unique effect of a unit increase in a covariate is multiplicative with respect to the hazard rate.,,A regression analysis method for survival analysis where the unique effect of a unit increase in a covariate is multiplicative with respect to the hazard rate.,A regression analysis for survival analysis where the unique effect of a unit increase in a covariate is multiplicative with respect to the hazard rate.,https://en.wikipedia.org/wiki/Proportional_hazards_model,,,AIO:SurvivalAnalysis +AIO:FixedEffectsModel,AIO:MachineLearningSubset,FEM,,Fixed Effects Model,FALSE,0.21,A statistical model in which the model parameters are fixed or non-random quantities.,,A regression analysis model in which the model parameters are fixed or non-random quantities.,A regression analysis in which the model parameters are fixed or non-random quantities.,https://en.wikipedia.org/wiki/Fixed_effects_model|,,,AIO:RegressionAnalysis +AIO:RidgeRegression,AIO:MachineLearningSubset,,,Ridge Regression,FALSE,0.20,A method of estimating the coefficients of multiple regression models in scenarios where the independent variables are highly correlated.,,A regression analysis method that estimates the coefficients of multiple regression models in scenarios where the independent variables are highly correlated.,A regression analysis that estimates the coefficients of multiple regression models in scenarios where the independent variables are highly correlated.,https://en.wikipedia.org/wiki/Ridge_regression|,,,AIO:RegressionAnalysis +AIO:LogisticRegression,AIO:MachineLearningSubset,,,Logistic Regression,FALSE,0.12,A statistical model that estimates the probability of an event occurring by modeling the log-odds of the event as a linear combination of one or more independent variables.,,A regression analysis model that estimates the probability of an event occurring by modeling the log-odds of the event as a linear combination of one or more independent variables.,A regression analysis that estimates the probability of an event occurring by modeling the log-odds of the event as a linear combination of one or more independent variables.,https://en.wikipedia.org/wiki/Logistic_regression|,,,AIO:RegressionAnalysis +AIO:LinearRegression,AIO:MachineLearningSubset,,,Linear Regression,FALSE,0.23,A linear approach for modeling the relationship between a scalar response and one or more explanatory variables.,,A regression analysis model that is a linear approach for modeling the relationship between a scalar response and one or more explanatory variables.,A regression analysis that is a linear approach for modeling the relationship between a scalar response and one or more explanatory variables.,https://en.wikipedia.org/wiki/Linear_regression|,,,AIO:RegressionAnalysis +AIO:LassoRegression,AIO:MachineLearningSubset,,,Lasso Regression,FALSE,0.00,A regression analysis method that performs both variable selection and regularization to enhance prediction accuracy and interpretability.,,A regression analysis method that performs both variable selection and regularization to enhance prediction accuracy and interpretability.,A regression analysis that performs both variable selection and regularization to enhance prediction accuracy and interpretability.,https://en.wikipedia.org/wiki/Lasso_(statistics)|,,,AIO:RegressionAnalysis +AIO:GeneralizedLinearModel,AIO:MachineLearningSubset,GLM,,Generalized Linear Model,FALSE,0.07,A model that generalizes linear regression by relating the linear model to the response variable via a link function and allowing the variance of each measurement to be a function of its predicted value.,,A machine learning model that generalizes linear regression by relating the linear model to the response variable via a link function and allowing the variance of each measurement to be a function of its predicted value.,A regression analysis that relates the linear model to the response variable via a link function and allowing the variance of each measurement to be a function of its predicted value.,https://en.wikipedia.org/wiki/Generalized_linear_model|,,,AIO:RegressionAnalysis +AIO:SpatialRegression,AIO:MachineLearningSubset,,,Spatial Regression,FALSE,0.11,A regression method used to model spatial relationships.,,A regression analysis method used to model spatial relationships.,A regression analysis used to model spatial relationships.,https://gisgeography.com/spatial-regression-models-arcgis/|,,,AIO:RegressionAnalysis +AIO:RandomEffectsModel,AIO:MachineLearningSubset,REM,,Random Effects Model,FALSE,0.27,A statistical model where the model parameters are random variables.,,A regression analysis model where the model parameters are random variables.,A regression analysis where the model parameters are random variables.,https://en.wikipedia.org/wiki/Random_effects_model|,,,AIO:RegressionAnalysis +AIO:LeastsquaresAnalysis,AIO:MachineLearningSubset,,,Least-squares Analysis,FALSE,#N/A,,,A regression analysis which approximates the solution of overdetermined systems by minimizing the sum of the squares of the residuals.,A regression analysis which approximates the solution of overdetermined systems by minimizing the sum of the squares of the residuals.,https://en.wikipedia.org/wiki/Least_squares|,,,AIO:RegressionAnalysis +AIO:GaussianNoiseLayer,AIO:LayerSubset,,,GaussianNoise Layer,FALSE,0.89,"Apply additive zero-centered Gaussian noise. This is useful to mitigate overfitting (you could see it as a form of random data augmentation). Gaussian Noise (GS) is a natural choice as corruption process for real valued inputs. As it is a regularization layer, it is only active at training time.","Apply additive zero-centered Gaussian noise. This is useful to mitigate overfitting (you could see it as a form of random data augmentation). Gaussian Noise (GS) is a natural choice as corruption process for real valued inputs. As it is a regularization layer, it is only active at training time.",A regularization layer that applies additive zero-centered Gaussian noise.,A regularization layer that applies additive zero-centered Gaussian noise.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GaussianNoise,,,AIO:RegularizationLayer +AIO:AlphaDropoutLayer,AIO:LayerSubset,,,AlphaDropout Layer,FALSE,0.74,"Applies Alpha Dropout to the input. Alpha Dropout is a Dropout that keeps mean and variance of inputs to their original values, in order to ensure the self-normalizing property even after this dropout. Alpha Dropout fits well to Scaled Exponential Linear Units by randomly setting activations to the negative saturation value.","Applies Alpha Dropout to the input. Alpha Dropout is a Dropout that keeps mean and variance of inputs to their original values, in order to ensure the self-normalizing property even after this dropout. Alpha Dropout fits well to Scaled Exponential Linear Units by randomly setting activations to the negative saturation value.",A regularization layer that applies Alpha Dropout to the input keeping mean and variance of inputs to ensure self-normalizing property.,A regularization layer that applies Alpha Dropout to the input keeping mean and variance of inputs to ensure self-normalizing property.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/AlphaDropout,,,AIO:RegularizationLayer +AIO:ActivityRegularizationLayer,AIO:LayerSubset,,,ActivityRegularization Layer,FALSE,0.31,Layer that applies an update to the cost function based input activity.,,A regularization layer that applies an update to the cost function based on input activity.,A regularization layer that applies an update to the cost function based on input activity.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ActivityRegularization,,,AIO:RegularizationLayer +AIO:DropoutLayer,AIO:LayerSubset,,,Dropout Layer,FALSE,0.92,"Applies Dropout to the input. The Dropout layer randomly sets input units to 0 with a frequency of rate at each step during training time, which helps prevent overfitting. Inputs not set to 0 are scaled up by 1/(1 - rate) such that the sum over all inputs is unchanged. Note that the Dropout layer only applies when training is set to True such that no values are dropped during inference. When using model.fit, training will be appropriately set to True automatically, and in other contexts, you can set the kwarg explicitly to True when calling the layer. (This is in contrast to setting trainable=False for a Dropout layer. trainable does not affect the layer's behavior, as Dropout does not have any variables/weights that can be frozen during training.)","Applies Dropout to the input. The Dropout layer randomly sets input units to 0 with a frequency of rate at each step during training time, which helps prevent overfitting. Inputs not set to 0 are scaled up by 1/(1 - rate) such that the sum over all inputs is unchanged. Note that the Dropout layer only applies when training is set to True such that no values are dropped during inference. When using model.fit, training will be appropriately set to True automatically, and in other contexts, you can set the kwarg explicitly to True when calling the layer. (This is in contrast to setting trainable=False for a Dropout layer. trainable does not affect the layer's behavior, as Dropout does not have any variables/weights that can be frozen during training.)",A regularization layer that applies Dropout to the input,A regularization layer that applies Dropout to the input,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Dropout,,,AIO:RegularizationLayer +AIO:GaussianDropoutLayer,AIO:LayerSubset,,,GaussianDropout Layer,FALSE,0.75,"Apply multiplicative 1-centered Gaussian noise. As it is a regularization layer, it is only active at training time.","Apply multiplicative 1-centered Gaussian noise. As it is a regularization layer, it is only active at training time.",A regularization layer that applies multiplicative 1-centered Gaussian noise.,A regularization layer that applies multiplicative 1-centered Gaussian noise.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/GaussianDropout,,,AIO:RegularizationLayer +AIO:SpatialDropout1DLayer,AIO:LayerSubset,,,SpatialDropout1D Layer,FALSE,0.80,"Spatial 1D version of Dropout. This version performs the same function as Dropout, however, it drops entire 1D feature maps instead of individual elements. If adjacent frames within feature maps are strongly correlated (as is normally the case in early convolution layers) then regular dropout will not regularize the activations and will otherwise just result in an effective Learning rate decrease. In this case, SpatialDropout1D will help promote independence between feature maps and should be used instead.","Spatial 1D version of Dropout. This version performs the same function as Dropout, however, it drops entire 1D feature maps instead of individual elements. If adjacent frames within feature maps are strongly correlated (as is normally the case in early convolution layers) then regular dropout will not regularize the activations and will otherwise just result in an effective Learning rate decrease. In this case, SpatialDropout1D will help promote independence between feature maps and should be used instead.",A regularization layer that performs the same function as Dropout but drops entire 1D feature maps instead of individual elements.,A regularization layer that performs the same function as Dropout but drops entire 1D feature maps instead of individual elements.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/SpatialDropout1D,,,AIO:RegularizationLayer +AIO:SpatialDropout2DLayer,AIO:LayerSubset,,,SpatialDropout2D Layer,FALSE,0.80,"Spatial 2D version of Dropout. This version performs the same function as Dropout, however, it drops entire 2D feature maps instead of individual elements. If adjacent pixels within feature maps are strongly correlated (as is normally the case in early convolution layers) then regular dropout will not regularize the activations and will otherwise just result in an effective Learning rate decrease. In this case, SpatialDropout2D will help promote independence between feature maps and should be used instead.a","Spatial 2D version of Dropout. This version performs the same function as Dropout, however, it drops entire 2D feature maps instead of individual elements. If adjacent pixels within feature maps are strongly correlated (as is normally the case in early convolution layers) then regular dropout will not regularize the activations and will otherwise just result in an effective Learning rate decrease. In this case, SpatialDropout2D will help promote independence between feature maps and should be used instead.a",A regularization layer that performs the same function as Dropout but drops entire 2D feature maps instead of individual elements.,A regularization layer that performs the same function as Dropout but drops entire 2D feature maps instead of individual elements.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/SpatialDropout2D,,,AIO:RegularizationLayer +AIO:SpatialDropout3DLayer,AIO:LayerSubset,,,SpatialDropout3D Layer,FALSE,0.80,"Spatial 3D version of Dropout. This version performs the same function as Dropout, however, it drops entire 3D feature maps instead of individual elements. If adjacent voxels within feature maps are strongly correlated (as is normally the case in early convolution layers) then regular dropout will not regularize the activations and will otherwise just result in an effective Learning rate decrease. In this case, SpatialDropout3D will help promote independence between feature maps and should be used instead.","Spatial 3D version of Dropout. This version performs the same function as Dropout, however, it drops entire 3D feature maps instead of individual elements. If adjacent voxels within feature maps are strongly correlated (as is normally the case in early convolution layers) then regular dropout will not regularize the activations and will otherwise just result in an effective Learning rate decrease. In this case, SpatialDropout3D will help promote independence between feature maps and should be used instead.",A regularization layer that performs the same function as Dropout but drops entire 3D feature maps instead of individual elements.,A regularization layer that performs the same function as Dropout but drops entire 3D feature maps instead of individual elements.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/SpatialDropout3D,,,AIO:RegularizationLayer +AIO:SparseLearning,AIO:NetworkSubset,Sparse dictionary Learning|Sparse coding,,Sparse Learning,FALSE,0.39,Finding sparse representations of input data as a linear combination of basic elements and identifying those elements.,,A representation learning network that finds sparse representations of input data as a linear combination of basic elements and identifies those elements.,A representation learning network that finds sparse representations of input data as a linear combination of basic elements and identifies those elements.,https://en.wikipedia.org/wiki/Sparse_dictionary_learning,,,AIO:RepresentationLearning +AIO:Cropping1DLayer,AIO:LayerSubset,,,Cropping1D Layer,FALSE,0.55,Cropping layer for 1D input (e.g. temporal sequence). It crops along the time dimension (axis 1).,Cropping layer for 1D input (e.g. temporal sequence). It crops along the time dimension (axis 1).,A layer that crops along the time dimension (axis 1) for 1D input.,A reshaping layer that crops along the time dimension (axis 1) for 1D input.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Cropping1D,,,AIO:ReshapingLayer +AIO:FlattenLayer,AIO:LayerSubset,,,Flatten Layer,FALSE,0.92,Flattens the input. Does not affect the batch size.,Flattens the input. Does not affect the batch size.,A layer that flattens the input,A reshaping layer that flattens the input,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Flatten,,,AIO:ReshapingLayer +AIO:PermuteLayer,AIO:LayerSubset,,,Permute Layer,FALSE,0.55,Permutes the dimensions of the input according to a given pattern. Useful e.g. connecting RNNs and convnets.,Permutes the dimensions of the input according to a given pattern. Useful e.g. connecting RNNs and convnets.,A layer that permutes the dimensions of the input according to a given pattern.,A reshaping layer that permutes the dimensions of the input according to a given pattern.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Permute,,,AIO:ReshapingLayer +AIO:RepeatVectorLayer,AIO:LayerSubset,,,RepeatVector Layer,FALSE,0.56,Repeats the input n times.,Repeats the input n times.,A layer that repeats the input n times.,A reshaping layer that repeats the input n times.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RepeatVector,,,AIO:ReshapingLayer +AIO:ReshapeLayer,AIO:LayerSubset,,,Reshape Layer,FALSE,0.30,Layer that reshapes inputs into the given shape.,,A layer that reshapes the inputs into the given shape.,A reshaping layer that reshapes the inputs into the given shape.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Reshape,,,AIO:ReshapingLayer +AIO:UpSampling1DLayer,AIO:LayerSubset,,,UpSampling1D Layer,FALSE,0.52,Upsampling layer for 1D inputs. Repeats each temporal step size times along the time axis.,Upsampling layer for 1D inputs. Repeats each temporal step size times along the time axis.,A layer that upsamples the input by repeating each temporal step size times along the time axis.,A reshaping layer that upsamples the input by repeating each temporal step size times along the time axis.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/UpSampling1D,,,AIO:ReshapingLayer +AIO:ZeroPadding3DLayer,AIO:LayerSubset,,,ZeroPadding3D Layer,FALSE,0.93,Zero-padding layer for 3D data (spatial or spatio-temporal).,Zero-padding layer for 3D data (spatial or spatio-temporal).,A layer that zero-pads the input along the depth,A reshaping layer that zero-pads the input along the depth,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ZeroPadding3D,,,AIO:ReshapingLayer +AIO:ZeroPadding2DLayer,AIO:LayerSubset,,,ZeroPadding2D Layer,FALSE,0.88,"Zero-padding layer for 2D input (e.g. picture). This layer can add rows and columns of zeros at the top, bottom, left and right side of an image tensor.","Zero-padding layer for 2D input (e.g. picture). This layer can add rows and columns of zeros at the top, bottom, left and right side of an image tensor.",A layer that zero-pads the input along the height and width dimensions.,A reshaping layer that zero-pads the input along the height and width dimensions.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ZeroPadding2D,,,AIO:ReshapingLayer +AIO:ZeroPadding1DLayer,AIO:LayerSubset,,,ZeroPadding1D Layer,FALSE,0.87,Zero-padding layer for 1D input (e.g. temporal sequence).,Zero-padding layer for 1D input (e.g. temporal sequence).,A layer that zero-pads the input along the time axis.,A reshaping layer that zero-pads the input along the time axis.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ZeroPadding1D,,,AIO:ReshapingLayer +AIO:DataGenerationBias,AIO:BiasSubset,,,Data Generation Bias,FALSE,0.41,Bias from adding synthetic or redundant data samples to a dataset.,Bias from adding synthetic or redundant data samples to a dataset.,A selection and sampling bias arising from adding synthetic or redundant data samples to a dataset.,A selection and sampling bias arising from adding synthetic or redundant data samples to a dataset.,https://en.wikipedia.org/wiki/Selection_bias,,,AIO:SelectionAndSamplingBias +AIO:TemporalBias,AIO:BiasSubset,,,Temporal Bias,FALSE,0.36,Bias arising from differences in populations and behaviors over time.,,A selection and sampling bias arising from differences in populations and behaviors over time.,A selection and sampling bias arising from differences in populations and behaviors over time.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:SelectionAndSamplingBias +AIO:MeasurementBias,AIO:BiasSubset,,,Measurement Bias,FALSE,0.33,"Bias arising when features and labels are proxies for desired quantities, potentially leading to differential performance.",,A selection and sampling bias arising when features and labels are proxies for desired quantities potentially leading to differential performance.,A selection and sampling bias arising when features and labels are proxies for desired quantities potentially leading to differential performance.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:SelectionAndSamplingBias +AIO:EvaluationBias,AIO:BiasSubset,,,Evaluation Bias,FALSE,0.29,Bias arising when testing populations do not equally represent user populations or when inappropriate performance metrics are used.,,A selection and sampling bias arising when testing populations do not equally represent user populations or when inappropriate performance metrics are used.,A selection and sampling bias arising when testing populations do not equally represent user populations or when inappropriate performance metrics are used.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:SelectionAndSamplingBias +AIO:DetectionBias,AIO:BiasSubset,,,Detection Bias,FALSE,0.46,"Systematic differences between groups in how outcomes are determined, potentially over- or underestimating effect size.","Systematic differences between groups in how outcomes are determined, potentially over- or underestimating effect size.",A selection and sampling bias characterized by systematic differences between groups in how outcomes are determined potentially over- or underestimating effect size.,A selection and sampling bias characterized by systematic differences between groups in how outcomes are determined potentially over- or underestimating effect size.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:SelectionAndSamplingBias +AIO:PopulationBias,AIO:BiasSubset,,,Population Bias,FALSE,0.36,Systematic distortions in demographics or other user characteristics between represented users and the target population.,,A selection and sampling bias characterized by systematic distortions in demographics or other user characteristics between represented users and the target population.,A selection and sampling bias characterized by systematic distortions in demographics or other user characteristics between represented users and the target population.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:SelectionAndSamplingBias +AIO:RepresentationBias,AIO:BiasSubset,,,Representation Bias,FALSE,0.41,"Bias due to non-random sampling of subgroups, making trends non-generalizable to new populations.","Bias due to non-random sampling of subgroups, making trends non-generalizable to new populations.",A selection and sampling bias due to non-random sampling of subgroups making trends non-generalizable to new populations.,A selection and sampling bias due to non-random sampling of subgroups making trends non-generalizable to new populations.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:SelectionAndSamplingBias +AIO:UncertaintyBias,AIO:BiasSubset,,,Uncertainty Bias,FALSE,0.42,"Bias favoring groups better represented in training data, due to less prediction uncertainty.","Bias favoring groups better represented in training data, due to less prediction uncertainty.",A selection and sampling bias favoring groups better represented in training data due to less prediction uncertainty.,A selection and sampling bias favoring groups better represented in training data due to less prediction uncertainty.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:SelectionAndSamplingBias +AIO:EcologicalFallacyBias,AIO:BiasSubset,Ecological Fallacy,,Ecological Fallacy Bias,FALSE,0.32,Bias occurring when an inference about an individual is made based on their group membership.,,A selection and sampling bias occurring when an inference about an individual is made based on their group membership.,A selection and sampling bias occurring when an inference about an individual is made based on their group membership.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:SelectionAndSamplingBias +AIO:ExclusionBias,AIO:BiasSubset,,,Exclusion Bias,FALSE,0.28,Bias occurring when specific groups of user populations are excluded from testing and analysis.,,A selection and sampling bias occurring when specific groups of user populations are excluded from testing and analysis.,A selection and sampling bias occurring when specific groups of user populations are excluded from testing and analysis.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:SelectionAndSamplingBias +AIO:PopularityBias,AIO:BiasSubset,,,Popularity Bias,FALSE,0.44,"Selection bias where more popular items are more exposed, under-representing less popular items.","Selection bias where more popular items are more exposed, under-representing less popular items.",A selection and sampling bias where more popular items are more exposed under-representing less popular items.,A selection and sampling bias where more popular items are more exposed under-representing less popular items.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:SelectionAndSamplingBias +AIO:SimponsParadoxBias,AIO:BiasSubset,Simpson's Paradox,,Simpon's Paradox Bias,FALSE,#N/A,A statistical phenomenon where the association between two variables changes when controlling for another variable.,,A selection and sampling bias where the association between two variables changes when controlling for another variable.,A selection and sampling bias where the association between two variables changes when controlling for another variable.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:SelectionAndSamplingBias +AIO:RegressionAnalysis,AIO:MachineLearningSubset,Regression analysis|Regression model,,Regression Analysis,FALSE,0.00,A set of statistical processes for estimating the relationships between a dependent variable and one or more independent variables.,,A set of statistical processes for estimating the relationships between a dependent variable and one or more independent variables.,A set of statistical processes for estimating the relationships between a dependent variable and one or more independent variables.,https://en.wikipedia.org/wiki/Regression_analysis|,,,AIO:SupervisedLearning +AIO:Layer,AIO:ClassSubset,,,Layer,FALSE,0.00,A structure or network topology in a deep learning model that takes information from previous layers and passes it to the next layer.,,A structure or network topology in a deep learning model that takes information from previous layers and passes it to the next layer.,A structure or network topology in a deep learning model that takes information from previous layers and passes it to the next layer.,https://en.wikipedia.org/wiki/Layer_(deep_learning),,,owl:Thing +AIO:AssociationRuleLearning,AIO:MachineLearningSubset,,,Association Rule Learning,FALSE,0.32,A rule-based machine learning method for discovering interesting relations between variables in large databases.,,A supervised learning method focused on a rule-based approach for discovering interesting relations between variables in large databases.,A supervised learning focused on a rule-based approach for discovering interesting relations between variables in large databases.,https://en.wikipedia.org/wiki/Association_rule_learning|,,,AIO:SupervisedLearning +AIO:Classification,AIO:MachineLearningSubset,,,Classification,FALSE,0.44,"Methods that distinguish and distribute kinds of ""things"" into different groups.","Methods that distinguish and distribute kinds of ""things"" into different groups.","A supervised learning task focused on methods that distinguish and distribute kinds of ""things"" into different groups.","A supervised learning focused on methods that distinguish and distribute kinds of ""things"" into different groups.",https://en.wikipedia.org/wiki/Classification_(general_theory)|,,,AIO:SupervisedLearning +AIO:HopfieldNetwork,AIO:NetworkSubset,HN|Ising model of a neural network|Ising–Lenz–Little model,,Hopfield Network,FALSE,0.64,"A Hopfield network is a type of recurrent artificial neural network that serves as a content-addressable (""associative"") memory system. It uses binary threshold nodes or continuous variables to store and recall memory patterns, providing a model for understanding human memory.",,A symmetrically connected network that is a type of recurrent artificial neural network serving as a content-addressable memory system.,A symmetrically connected network that is a type of recurrent artificial neural network serving as a content-addressable memory system.,https://en.wikipedia.org/wiki/Hopfield_network,Layers: Backfed input,AIO:BackfedInputLayer,AIO:SymmetricallyConnectedNetwork +AIO:BoltzmannMachineNetwork,AIO:NetworkSubset,BM|stochastic Hopfield network with hidden units|Sherrington–Kirkpatrick model with external field|stochastic Ising-Lenz-Little model,,Boltzmann Machine Network,FALSE,0.57,"A Boltzmann machine is a type of stochastic recurrent neural network and Markov random field, translated from statistical physics for use in cognitive science.",,A symmetrically connected network that is a type of stochastic recurrent neural network and Markov random field.,A symmetrically connected network that is a type of stochastic recurrent neural network and Markov random field.,https://en.wikipedia.org/wiki/Boltzmann_machine,"Layers: Backfed Input, Probabilistic Hidden",AIO:BackfedInputLayer|AIO:ProbabilisticHiddenLayer,AIO:SymmetricallyConnectedNetwork +AIO:Network,AIO:ClassSubset,,,Network,FALSE,0.29,"A system of interconnected nodes or entities for communication, computation, or data exchange.",,A system of interconnected nodes or entities for communication computation or data exchange.,A system of interconnected nodes or entities for communication computation or data exchange.,,,,owl:Thing +AIO:Bias,AIO:BiasSubset,,,Bias,FALSE,0.18,Systematic error introduced into sampling or testing by selecting or encouraging one outcome or answer over others.,,A systematic error introduced into sampling or testing by selecting or encouraging one outcome or answer over others.,A systematic error introduced into sampling or testing by selecting or encouraging one outcome or answer over others.,https://www.merriam-webster.com/dictionary/bias,,,owl:Thing +AIO:TextVectorizationLayer,AIO:LayerSubset,,,TextVectorization Layer,FALSE,0.18,A preprocessing layer which maps text features to integer sequences.,,A preprocessing layer that maps text features to integer sequences.,A text preprocessing layer that maps text features to integer sequences.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/TextVectorization,,,AIO:TextPreprocessingLayer +AIO:CurriculumLearning,AIO:PreprocessingSubset,Sequential Learning|Structured Learning,Sequential learning|Complexity grading,Curriculum Learning,FALSE,0.13,"A training strategy in machine learning where models are trained on data in a meaningful order, starting with simpler examples and gradually increasing the complexity, to improve learning efficiency and model performance.",,A training strategy in machine learning where models are trained on data in a meaningful order starting with simpler examples and gradually increasing the complexity to improve learning efficiency and model performance.,A training strategy in machine learning where models are trained on data in a meaningful order starting with simpler examples and gradually increasing the complexity to improve learning efficiency and model performance.,,,,AIO:TrainingStrategies +AIO:KnowledgeTransfer,AIO:PreprocessingSubset,Skill Acquisition|Inductive Transfer,Pretrained models|Adaptation,Knowledge Transfer,FALSE,0.32,"The process by which knowledge is passed from one entity, such as a person, organization, or system, to another, facilitating learning and adaptation in the receiving entity through various methods such as teaching, training, or data exchange.",,The process by which knowledge is passed from one entity such as a person organization or system to another facilitating learning and adaptation in the receiving entity through various methods such as teaching training or data exchange.,A training strategy in which knowledge is passed from one entity such as a person organization or system to another facilitating learning and adaptation in the receiving entity through various methods such as teaching training or data exchange.,https://doi.org/10.1016/j.knosys.2015.01.010|,,,AIO:TrainingStrategies +AIO:BidirectionalTransformerLanguageModel,AIO:ModelSubset,Bidirectional Transformer LM|BERT,,Bidirectional Transformer Language Model,FALSE,0.19,"A bidirectional transformer language model, such as BERT, uses the transformer architecture to build deep bidirectional representations by predicting masked tokens based on their context.",,A transformer language model such as BERT that uses the transformer architecture to build deep bidirectional representations by predicting masked tokens based on their context.,A transformer language model such as BERT that uses the transformer architecture to build deep bidirectional representations by predicting masked tokens based on their context.,https://arxiv.org/abs/1810.04805|https://en.wikipedia.org/wiki/BERT_(language_model)|,,,AIO:TransformerLanguageModel +AIO:TransformerLLM,AIO:ModelSubset,Transformer Large Language Model,,Transformer LLM,FALSE,0.15,"A transformer LLM is a neural network model with large training corpuses and large sets of parameters that uses the transformer architecture based on multi-head attention mechanisms, allowing it to contextualize tokens within a context window for effective language understanding and generation.",,A transformer language model with large training corpuses and sets of parameters that uses the transformer architecture based on multi-head attention mechanisms allowing it to contextualize tokens within a context window for effective language understanding and generation.,A transformer language model with large training corpuses and sets of parameters that uses the transformer architecture based on multi-head attention mechanisms allowing it to contextualize tokens within a context window for effective language understanding and generation.,https://en.wikipedia.org/wiki/Transformer_(deep_learning_architecture),,,AIO:TransformerLanguageModel +AIO:MultimodalTransformer,AIO:ModelSubset,,vision-language model|unified encoder,Multimodal Transformer,FALSE,0.39,"A multimodal transformer processes and relates information from different modalities, such as text, images, and audio. It uses a shared embedding space and attention mechanism to learn joint representations across modalities.",,A transformer network that processes and relates information from different modalities such as text images and audio using a shared embedding space and attention mechanism to learn joint representations across modalities.,A transformer network that processes and relates information from different modalities such as text images and audio using a shared embedding space and attention mechanism to learn joint representations across modalities.,,,,AIO:TransformerNetwork +AIO:ContentProductionBias,AIO:BiasSubset,,,Content Production Bias,FALSE,0.63,"Bias from structural, lexical, semantic, and syntactic differences in user-generated content.","Bias from structural, lexical, semantic, and syntactic differences in user-generated content.",A use and interpretation bias arising from structural lexical semantic and syntactic differences in user-generated content.,A use and interpretation bias arising from structural lexical semantic and syntactic differences in user-generated content.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:UseAndInterpretationBias +AIO:LinkingBias,AIO:BiasSubset,,,Linking Bias,FALSE,0.45,"Bias arising when network attributes obtained from user connections, activities, or interactions misrepresent true user behavior.","Bias arising when network attributes obtained from user connections, activities, or interactions misrepresent true user behavior.",A use and interpretation bias arising when network attributes obtained from user connections activities or interactions misrepresent true user behavior.,A use and interpretation bias arising when network attributes obtained from user connections activities or interactions misrepresent true user behavior.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:UseAndInterpretationBias +AIO:ConceptDriftBias,AIO:BiasSubset,Concept Drift,,Concept Drift Bias,FALSE,0.23,"Bias due to the use of a system outside its planned domain of application, causing performance gaps between laboratory settings and the real world.",,A use and interpretation bias due to the use of a system outside its planned domain of application causing performance gaps between laboratory settings and the real world.,A use and interpretation bias due to the use of a system outside its planned domain of application causing performance gaps between laboratory settings and the real world.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:UseAndInterpretationBias +AIO:FeedbackLoopBias,AIO:BiasSubset,,,Feedback Loop Bias,FALSE,0.25,Effects occurring when an algorithm learns from user behavior and feeds that behavior back into the model.,,A use and interpretation bias occurring when an algorithm learns from user behavior and feeds that behavior back into the model.,A use and interpretation bias occurring when an algorithm learns from user behavior and feeds that behavior back into the model.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:UseAndInterpretationBias +AIO:ActivityBias,AIO:BiasSubset,,,Activity Bias,FALSE,0.29,"Selection bias occurring when systems/platforms get training data from their most active users, rather than less active or inactive users.",,A use and interpretation bias occurring when systems/platforms get training data from their most active users rather than less active or inactive users.,A use and interpretation bias occurring when systems/platforms get training data from their most active users rather than less active or inactive users.,https://en.wikipedia.org/wiki/Interpretive_bias,,,AIO:UseAndInterpretationBias +AIO:EmergentBias,AIO:BiasSubset,,,Emergent Bias,FALSE,0.24,Bias resulting from the use and reliance on algorithms across new or unanticipated contexts.,,A use and interpretation bias resulting from the use and reliance on algorithms across new or unanticipated contexts.,A use and interpretation bias resulting from the use and reliance on algorithms across new or unanticipated contexts.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:UseAndInterpretationBias +AIO:HostileAttributionBias,AIO:BiasSubset,,,Hostile Attribution Bias,FALSE,0.40,Bias where individuals perceive benign or ambiguous behaviors as hostile.,Bias where individuals perceive benign or ambiguous behaviors as hostile.,A use and interpretation bias where individuals perceive benign or ambiguous behaviors as hostile.,A use and interpretation bias where individuals perceive benign or ambiguous behaviors as hostile.,https://en.wikipedia.org/wiki/Interpretive_bias,,,AIO:UseAndInterpretationBias +AIO:DataDredgingBias,AIO:BiasSubset,Data Dredging,,Data Dredging Bias,FALSE,0.22,Statistical bias where testing many hypotheses in a dataset may yield apparent statistical significance even when results are nonsignificant.,,A use and interpretation bias where testing many hypotheses in a dataset may yield apparent statistical significance even when results are nonsignificant.,A use and interpretation bias where testing many hypotheses in a dataset may yield apparent statistical significance even when results are nonsignificant.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:UseAndInterpretationBias +AIO:word2vecSkipGram,AIO:NetworkSubset,W2V-SkipGram,SkipGram,word2vec-SkipGram,FALSE,#N/A,,,"A word2vec that predicts surrounding context words from the current word, giving more weight to nearby context words than distant ones.","A word2vec that predicts surrounding context words from the current word, giving more weight to nearby context words than distant ones.",https://en.wikipedia.org/wiki/Word2vec,"Layers: Input, Hidden, Output",AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:word2vec +AIO:word2vecCBOW,AIO:NetworkSubset,W2V-CBOW,CBOW,word2vec-CBOW,FALSE,#N/A,,,"A word2vec that predicts the current word from a window of surrounding context words, ignoring the order of context words.","A word2vec that predicts the current word from a window of surrounding context words, ignoring the order of context words.",https://en.wikipedia.org/wiki/Word2vec,"Layers: Input, Hidden, Output",AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:word2vec +AIO:Model,AIO:ModelSubset,,,Model,FALSE,0.38,"A model is an abstract representation of a complex system, generally assembled as a set of logical, mathematical, or conceptual properties to simulate or understand the system's behavior.",,An abstract representation of a complex system generally assembled as a set of logical mathematical or conceptual properties to simulate or understand the system's behavior.,An abstract representation of a complex system generally assembled as a set of logical mathematical or conceptual properties to simulate or understand the system's behavior.,https://en.wikipedia.org/wiki/Mathematical_model|,,,owl:Thing +AIO:PReLULayer,AIO:LayerSubset,,,PReLU Layer,FALSE,1.00,Parametric Rectified Linear Unit.,Parametric Rectified Linear Unit.,An activation layer that applies parametric rectified linear unit function element-wise.,An activation layer that applies parametric rectified linear unit function element-wise.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/PReLU,,,AIO:ActivationLayer +AIO:ELULayer,AIO:LayerSubset,,,ELU Layer,FALSE,0.85,Exponential Linear Unit.,Exponential Linear Unit.,An activation layer that applies the Exponential Linear Unit (ELU) function element-wise.,An activation layer that applies the Exponential Linear Unit (ELU) function element-wise.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ELU,,,AIO:ActivationLayer +AIO:LeakyReLULayer,AIO:LayerSubset,,,LeakyReLU Layer,FALSE,1.00,Leaky version of a Rectified Linear Unit.,Leaky version of a Rectified Linear Unit.,An activation layer that applies the leaky rectified linear unit function element-wise.,An activation layer that applies the leaky rectified linear unit function element-wise.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/LeakyReLU,,,AIO:ActivationLayer +AIO:ReLULayer,AIO:LayerSubset,,,ReLU Layer,FALSE,0.96,"Rectified Linear Unit activation function. With default values, it returns element-wise max(x, 0).","Rectified Linear Unit activation function. With default values, it returns element-wise max(x, 0).",An activation layer that applies the rectified linear unit function element-wise.,An activation layer that applies the rectified linear unit function element-wise.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ReLU,,,AIO:ActivationLayer +AIO:SoftmaxLayer,AIO:LayerSubset,,,Softmax Layer,FALSE,0.92,Softmax activation function.,Softmax activation function.,An activation layer that applies the softmax function to the inputs.,An activation layer that applies the softmax function to the inputs.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Softmax,,,AIO:ActivationLayer +AIO:ThresholdedReLULayer,AIO:LayerSubset,,,ThresholdedReLU Layer,FALSE,1.00,Thresholded Rectified Linear Unit.,Thresholded Rectified Linear Unit.,An activation layer that applies the thresholded rectified linear unit function element-wise.,An activation layer that applies the thresholded rectified linear unit function element-wise.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/ThresholdedReLU,,,AIO:ActivationLayer +AIO:RankingBias,AIO:BiasSubset,,,Ranking Bias,FALSE,0.33,"The idea that top-ranked results are the most relevant and important, leading to more clicks than other results.",,An anchoring bias characterized by the idea that top-ranked results are the most relevant and important leading to more clicks than other results.,An anchoring bias characterized by the idea that top-ranked results are the most relevant and important leading to more clicks than other results.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:AnchoringBias +AIO:DeepNeuralNetwork,AIO:NetworkSubset,DNN,,Deep Neural Network,FALSE,0.81,"A deep neural network (DNN) is a type of artificial neural network (ANN) characterized by multiple hidden layers between the input and output layers. Each layer consists of interconnected neurons that process and transmit information. DNNs can model complex patterns and representations in data through their hierarchical structure, where each layer extracts increasingly abstract features from the input. DNNs are widely used in various applications, including image and speech recognition, natural language processing, and more, due to their ability to learn and generalize from large amounts of data.","A deep neural network (DNN) is a type of artificial neural network (ANN) characterized by multiple hidden layers between the input and output layers. Each layer consists of interconnected neurons that process and transmit information. DNNs can model complex patterns and representations in data through their hierarchical structure, where each layer extracts increasingly abstract features from the input. DNNs are widely used in various applications, including image and speech recognition, natural language processing, and more, due to their ability to learn and generalize from large amounts of data.",An artificial neural network characterized by multiple hidden layers between the input and output layers.,An artificial neural network characterized by multiple hidden layers between the input and output layers.,,,,AIO:ArtificialNeuralNetwork +AIO:FeedbackNetwork,AIO:NetworkSubset,FBN,,Feedback Network,FALSE,0.35,A feedback network iteratively refines its representations based on feedback from previous iterations' outputs.,,An artificial neural network that refines its representations iteratively based on feedback from previous outputs.,An artificial neural network that refines its representations iteratively based on feedback from previous outputs.,,"Layers: Input, Hidden, Output, Hidden",AIO:InputLayer|AIO:HiddenLayer|AIO:OutputLayer,AIO:ArtificialNeuralNetwork +AIO:Perceptron,AIO:NetworkSubset,Single Layer Perceptron|SLP|Feed-Forward Network|FFN,,Perceptron,FALSE,0.71,"A perceptron is a supervised learning algorithm for binary classification, deciding if an input belongs to a class using a linear predictor function that combines weights with the feature vector.",,An artificial neural network with a supervised learning algorithm for binary classification using a linear predictor function.,An artificial neural network with a supervised learning algorithm for binary classification using a linear predictor function.,,"Layers: Input, Output",AIO:InputLayer|AIO:OutputLayer,AIO:ArtificialNeuralNetwork +AIO:MultiHeadAttentionLayer,AIO:LayerSubset,,,MultiHeadAttention Layer,FALSE,0.96,"MultiHeadAttention layer. This is an implementation of multi-headed attention as described in the paper ""Attention is all you Need"" (Vaswani et al., 2017). If query, key, value are the same, then this is self-attention. Each timestep in query attends to the corresponding sequence in key, and returns a fixed-width vector.This layer first projects query, key and value. These are (effectively) a list of tensors of length num_attention_heads, where the corresponding shapes are (batch_size, , key_dim), (batch_size, , key_dim), (batch_size, , value_dim).Then, the query and key tensors are dot-producted and scaled. These are softmaxed to obtain attention probabilities. The value tensors are then interpolated by these probabilities, then concatenated back to a single tensor. Finally, the result tensor with the last dimension as value_dim can take an linear projection and return. When using MultiHeadAttention inside a custom Layer, the custom Layer must implement build() and call MultiHeadAttention's _build_from_signature(). This enables weights to be restored correctly when the model is loaded.","MultiHeadAttention layer. This is an implementation of multi-headed attention as described in the paper ""Attention is all you Need"" (Vaswani et al., 2017). If query, key, value are the same, then this is self-attention. Each timestep in query attends to the corresponding sequence in key, and returns a fixed-width vector.This layer first projects query, key and value. These are (effectively) a list of tensors of length num_attention_heads, where the corresponding shapes are (batch_size, , key_dim), (batch_size, , key_dim), (batch_size, , value_dim).Then, the query and key tensors are dot-producted and scaled. These are softmaxed to obtain attention probabilities. The value tensors are then interpolated by these probabilities, then concatenated back to a single tensor. Finally, the result tensor with the last dimension as value_dim can take an linear projection and return. When using MultiHeadAttention inside a custom Layer, the custom Layer must implement build() and call MultiHeadAttention's _build_from_signature(). This enables weights to be restored correctly when the model is loaded.",An attention layer that allows the model to attend to information from different representation subspaces.,An attention layer that allows the model to attend to information from different representation subspaces.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/MultiHeadAttention,,,AIO:AttentionLayer +AIO:AdditiveAttentionLayer,AIO:LayerSubset,,,AdditiveAttention Layer,FALSE,0.79,"Additive attention layer, a.k.a. Bahdanau-style attention.","Additive attention layer, a.k.a. Bahdanau-style attention.",An attention layer that implements additive attention also known as Bahdanau-style attention.,An attention layer that implements additive attention also known as Bahdanau-style attention.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/AdditiveAttention,,,AIO:AttentionLayer +AIO:VariationalAutoEncoder,AIO:NetworkSubset,VAE,,Variational Auto Encoder,FALSE,0.84,"A variational autoencoder (VAE) is a type of artificial neural network used for unsupervised learning. It consists of an encoder, which maps input data to a latent space, and a decoder, which reconstructs the input data from the latent space. Unlike traditional autoencoders, VAEs impose a probabilistic structure on the latent space, enabling them to generate new data samples by sampling from the learned latent distribution. This probabilistic approach allows VAEs to learn smooth and meaningful latent representations, making them useful for tasks such as data generation, anomaly detection, and semi-supervised learning.",,An autoencoder network that imposes a probabilistic structure on the latent space for unsupervised learning.,An autoencoder network that imposes a probabilistic structure on the latent space for unsupervised learning.,,"Layers: Input, Probabilistic Hidden, Matched Output-Input",AIO:InputLayer|AIO:ProbabilisticHiddenLayer|AIO:MatchedInputOutputLayer,AIO:AutoEncoderNetwork +AIO:DeepConvolutionalInverseGraphicsNetwork,AIO:NetworkSubset,DCIGN,,Deep Convolutional Inverse Graphics Network,FALSE,0.59,A deep convolutional inverse graphics network (DC-IGN) learns interpretable image representations disentangled for transformations like out-of-plane rotations and lighting variations. It consists of convolution and de-convolution layers and is trained using the stochastic gradient variational Bayes (SGVB) algorithm.,,An autoencoder network that learns interpretable disentangled image representations through convolution and de-convolution layers trained with the stochastic gradient variational Bayes algorithm.,An autoencoder network that learns interpretable disentangled image representations through convolution and de-convolution layers trained with the stochastic gradient variational Bayes algorithm.,,"Layers: Input, Kernel, Convolutional/Pool, Probabilistic Hidden, Convolutional/Pool, Kernel, Output",AIO:InputLayer|AIO:KernelLayer|AIO:ConvolutionalLayer|AIO:PoolingLayer|AIO:ProbabilisticHiddenLayer|AIO:OutputLayer,AIO:AutoEncoderNetwork +AIO:DenoisingAutoEncoder,AIO:NetworkSubset,DAE|Denoising Autoencoder,,Denoising Auto Encoder,FALSE,0.54,"Denoising autoencoders (DAEs) are neural networks trained to reconstruct the original undistorted input from a partially corrupted input, aiming to clean or denoise the corrupted input.",,An autoencoder network trained to reconstruct the original undistorted input from a partially corrupted input.,An autoencoder network trained to reconstruct the original undistorted input from a partially corrupted input.,https://doi.org/10.1145/1390156.1390294,"Layers: Noisy Input, Hidden, Matched Output-Input",AIO:NoisyInputLayer|AIO:HiddenLayer|AIO:MatchedInputOutputLayer,AIO:AutoEncoderNetwork +AIO:SparseAutoEncoder,AIO:NetworkSubset,SAE|Sparse AE|Sparse Autoencoder,,Sparse Auto Encoder,FALSE,0.66,"Sparse autoencoders have more hidden units than inputs but constrain only a few hidden units to be active at once, forcing the model to capture unique statistical features of the training data.",,An autoencoder network with more hidden units than inputs that constrains only a few hidden units to be active at once.,An autoencoder network with more hidden units than inputs that constrains only a few hidden units to be active at once.,,"Layers: Input, Hidden, Matched Output-Input",AIO:InputLayer|AIO:HiddenLayer|AIO:MatchedInputOutputLayer,AIO:AutoEncoderNetwork +AIO:RandomForest,AIO:MachineLearningSubset,,,Random Forest,FALSE,0.19,"An ensemble learning method for classification, regression, and other tasks that constructs a multitude of decision trees during training.",,An ensemble learning method for classification regression and other tasks that constructs a multitude of decision trees during training.,An ensemble learning method for classification regression and other tasks that constructs a multitude of decision trees during training.,https://en.wikipedia.org/wiki/Random_forest,,,AIO:EnsembleLearning +AIO:CenterCropLayer,AIO:LayerSubset,,,CenterCrop Layer,FALSE,0.77,"A preprocessing layer which crops images. This layers crops the central portion of the images to a target size. If an image is smaller than the target size, it will be resized and cropped so as to return the largest possible window in the image that matches the target aspect ratio. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats.","A preprocessing layer which crops images. This layers crops the central portion of the images to a target size. If an image is smaller than the target size, it will be resized and cropped so as to return the largest possible window in the image that matches the target aspect ratio. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats.",An image preprocessing layer that crops the central portion of images to a target size.,An image preprocessing layer that crops the central portion of images to a target size.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/CenterCrop,,,AIO:ImagePreprocessingLayer +AIO:RandomBrightnessLayer,AIO:LayerSubset,,,RandomBrightness Layer,FALSE,0.81,"A preprocessing layer which randomly adjusts brightness during training. This layer will randomly increase/reduce the brightness for the input RGB images. At inference time, the output will be identical to the input. Call the layer with training=True to adjust the brightness of the input. Note that different brightness adjustment factors will be apply to each the images in the batch.","A preprocessing layer which randomly adjusts brightness during training. This layer will randomly increase/reduce the brightness for the input RGB images. At inference time, the output will be identical to the input. Call the layer with training=True to adjust the brightness of the input. Note that different brightness adjustment factors will be apply to each the images in the batch.",An image preprocessing layer that randomly adjusts brightness during training.,An image preprocessing layer that randomly adjusts brightness during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomBrightness,,,AIO:ImagePreprocessingLayer +AIO:RandomContrastLayer,AIO:LayerSubset,,,RandomContrast Layer,FALSE,0.90,"A preprocessing layer which randomly adjusts contrast during training. This layer will randomly adjust the contrast of an image or images by a random factor. Contrast is adjusted independently for each channel of each image during training. For each channel, this layer computes the mean of the image pixels in the channel and then adjusts each component x of each pixel to (x - mean) * contrast_factor + mean. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and in integer or floating point dtype. By default, the layer will output floats. The output value will be clipped to the range [0, 255], the valid range of RGB colors.","A preprocessing layer which randomly adjusts contrast during training. This layer will randomly adjust the contrast of an image or images by a random factor. Contrast is adjusted independently for each channel of each image during training. For each channel, this layer computes the mean of the image pixels in the channel and then adjusts each component x of each pixel to (x - mean) * contrast_factor + mean. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and in integer or floating point dtype. By default, the layer will output floats. The output value will be clipped to the range [0, 255], the valid range of RGB colors.",An image preprocessing layer that randomly adjusts contrast during training.,An image preprocessing layer that randomly adjusts contrast during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomContrast,,,AIO:ImagePreprocessingLayer +AIO:RandomCropLayer,AIO:LayerSubset,,,RandomCrop Layer,FALSE,0.90,"A preprocessing layer which randomly crops images during training. During training, this layer will randomly choose a location to crop images down to a target size. The layer will crop all the images in the same batch to the same cropping location. At inference time, and during training if an input image is smaller than the target size, the input will be resized and cropped so as to return the largest possible window in the image that matches the target aspect ratio. If you need to apply random cropping at inference time, set training to True when calling the layer. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats.","A preprocessing layer which randomly crops images during training. During training, this layer will randomly choose a location to crop images down to a target size. The layer will crop all the images in the same batch to the same cropping location. At inference time, and during training if an input image is smaller than the target size, the input will be resized and cropped so as to return the largest possible window in the image that matches the target aspect ratio. If you need to apply random cropping at inference time, set training to True when calling the layer. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats.",An image preprocessing layer that randomly crops images during training.,An image preprocessing layer that randomly crops images during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomCrop,,,AIO:ImagePreprocessingLayer +AIO:RandomFlipLayer,AIO:LayerSubset,,,RandomFlip Layer,FALSE,0.87,"A preprocessing layer which randomly flips images during training. This layer will flip the images horizontally and or vertically based on the mode attribute. During inference time, the output will be identical to input. Call the layer with training=True to flip the input. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats.","A preprocessing layer which randomly flips images during training. This layer will flip the images horizontally and or vertically based on the mode attribute. During inference time, the output will be identical to input. Call the layer with training=True to flip the input. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats.",An image preprocessing layer that randomly flips images during training.,An image preprocessing layer that randomly flips images during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomFlip,,,AIO:ImagePreprocessingLayer +AIO:RandomRotationLayer,AIO:LayerSubset,,,RandomRotation Layer,FALSE,0.42,A preprocessing layer which randomly rotates images during training.,A preprocessing layer which randomly rotates images during training.,An image preprocessing layer that randomly rotates images during training.,An image preprocessing layer that randomly rotates images during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomRotation,,,AIO:ImagePreprocessingLayer +AIO:RandomTranslationLayer,AIO:LayerSubset,,,RandomTranslation Layer,FALSE,0.84,"A preprocessing layer which randomly translates images during training. This layer will apply random translations to each image during training, filling empty space according to fill_mode. aInput pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats.","A preprocessing layer which randomly translates images during training. This layer will apply random translations to each image during training, filling empty space according to fill_mode. aInput pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats.",An image preprocessing layer that randomly translates images during training.,An image preprocessing layer that randomly translates images during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomTranslation,,,AIO:ImagePreprocessingLayer +AIO:RandomHeightLayer,AIO:LayerSubset,,,RandomHeight Layer,FALSE,0.87,"A preprocessing layer which randomly varies image height during training. This layer adjusts the height of a batch of images by a random factor. The input should be a 3D (unbatched) or 4D (batched) tensor in the ""channels_last"" image data format. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. By default, this layer is inactive during inference.","A preprocessing layer which randomly varies image height during training. This layer adjusts the height of a batch of images by a random factor. The input should be a 3D (unbatched) or 4D (batched) tensor in the ""channels_last"" image data format. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. By default, this layer is inactive during inference.",An image preprocessing layer that randomly varies image height during training.,An image preprocessing layer that randomly varies image height during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomHeight,,,AIO:ImagePreprocessingLayer +AIO:RandomWidthLayer,AIO:LayerSubset,,,RandomWidth Layer,FALSE,0.87,"A preprocessing layer which randomly varies image width during training. This layer will randomly adjusts the width of a batch of images of a batch of images by a random factor. The input should be a 3D (unbatched) or 4D (batched) tensor in the ""channels_last"" image data format. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. By default, this layer is inactive during inference.","A preprocessing layer which randomly varies image width during training. This layer will randomly adjusts the width of a batch of images of a batch of images by a random factor. The input should be a 3D (unbatched) or 4D (batched) tensor in the ""channels_last"" image data format. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. By default, this layer is inactive during inference.",An image preprocessing layer that randomly varies image width during training.,An image preprocessing layer that randomly varies image width during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomWidth,,,AIO:ImagePreprocessingLayer +AIO:RandomZoomLayer,AIO:LayerSubset,,,RandomZoom Layer,FALSE,0.76,"A preprocessing layer which randomly zooms images during training. This layer will randomly zoom in or out on each axis of an image independently, filling empty space according to fill_mode.Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats.","A preprocessing layer which randomly zooms images during training. This layer will randomly zoom in or out on each axis of an image independently, filling empty space according to fill_mode.Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats.",An image preprocessing layer that randomly zooms in or out on images during training.,An image preprocessing layer that randomly zooms in or out on images during training.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomZoom,,,AIO:ImagePreprocessingLayer +AIO:RescalingLayer,AIO:LayerSubset,,,Rescaling Layer,FALSE,0.17,A preprocessing layer which rescales input values to a new range.,,A preprocessing layer that rescales input values to a new range.,An image preprocessing layer that rescales input values to a new range.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Rescaling,,,AIO:ImagePreprocessingLayer +AIO:ResizingLayer,AIO:LayerSubset,,,Resizing Layer,FALSE,0.86,"A preprocessing layer which resizes images. This layer resizes an image input to a target height and width. The input should be a 4D (batched) or 3D (unbatched) tensor in ""channels_last"" format. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. This layer can be called on tf.RaggedTensor batches of input images of distinct sizes, and will resize the outputs to dense tensors of uniform size.","A preprocessing layer which resizes images. This layer resizes an image input to a target height and width. The input should be a 4D (batched) or 3D (unbatched) tensor in ""channels_last"" format. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. This layer can be called on tf.RaggedTensor batches of input images of distinct sizes, and will resize the outputs to dense tensors of uniform size.",A preprocessing layer that resizes images to a target size.,An image preprocessing layer that resizes images to a target size.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/Resizing,,,AIO:ImagePreprocessingLayer +AIO:PresentationBias,AIO:BiasSubset,,,Presentation Bias,FALSE,0.36,"Bias arising from how information is presented on the Web, via a user interface, due to rating or ranking of output, or through users' self-selected, biased interaction.",,An individual bias arising from how information is presented on the Web via a user interface due to rating or ranking of output or through users' self-selected biased interaction.,An individual bias arising from how information is presented on the Web via a user interface due to rating or ranking of output or through users' self-selected biased interaction.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:UserInteractionBias,AIO:BiasSubset,,,User Interaction Bias,FALSE,0.45,"Bias arising when a user imposes their own biases during interaction with data, output, results, etc.","Bias arising when a user imposes their own biases during interaction with data, output, results, etc.",An individual bias arising when a user imposes their own biases during interaction with data output results etc.,An individual bias arising when a user imposes their own biases during interaction with data output results etc.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:ConsumerBias,AIO:BiasSubset,,,Consumer Bias,FALSE,0.20,"Bias arising when an algorithm or platform provides users a venue to express their biases, occurring from either side in a digital interaction.",,A bias arising when an algorithm or platform provides users a venue to express their biases occurring from either side in a digital interaction.,An individual bias arising when an algorithm or platform provides users a venue to express their biases occurring from either side in a digital interaction.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:AvailabilityHeuristicBias,AIO:BiasSubset,Availability Heuristic|Availability Bias,,Availability Heuristic Bias,FALSE,0.28,A mental shortcut where easily recalled information is overweighted in judgment and decision-making.,,A cognitive bias characterized by a mental shortcut where easily recalled information is overweighted in judgment and decision-making.,An individual bias characterized by a mental shortcut where easily recalled information is overweighted in judgment and decision-making.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:CognitiveBias,AIO:BiasSubset,,,Cognitive Bias,FALSE,0.29,"Systematic deviation from rational judgment and decision-making, including adaptive mental shortcuts known as heuristics.",,A systematic deviation from rational judgment and decision-making including adaptive mental shortcuts known as heuristics.,An individual bias characterized by deviations from rational judgment and decision-making including adaptive mental shortcuts known as heuristics.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:RashomonEffectBias,AIO:BiasSubset,Rashomon Effect|Rashomon Principle,,Rashomon Effect Bias,FALSE,0.54,"Differences in perspective, memory, recall, interpretation, and reporting of the same event by multiple persons or witnesses.","Differences in perspective, memory, recall, interpretation, and reporting of the same event by multiple persons or witnesses.",An individual bias characterized by differences in perspective memory recall interpretation and reporting of the same event by multiple persons or witnesses.,An individual bias characterized by differences in perspective memory recall interpretation and reporting of the same event by multiple persons or witnesses.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:AutomationComplacencyBias,AIO:BiasSubset,Automation Complaceny,,Automation Complacency Bias,FALSE,0.73,"Over-reliance on automated systems, leading to attenuated human skills, such as with spelling and autocorrect.","Over-reliance on automated systems, leading to attenuated human skills, such as with spelling and autocorrect.",A bias characterized by over-reliance on automated systems leading to attenuated human skills.,An individual bias characterized by over-reliance on automated systems leading to attenuated human skills.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:BehavioralBias,AIO:BiasSubset,,,Behavioral Bias,FALSE,0.45,"Systematic distortions in user behavior across platforms or contexts, or across users represented in different datasets.","Systematic distortions in user behavior across platforms or contexts, or across users represented in different datasets.",An individual bias characterized by systematic distortions in user behavior across platforms or contexts or across users represented in different datasets.,An individual bias characterized by systematic distortions in user behavior across platforms or contexts or across users represented in different datasets.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:AnchoringBias,AIO:BiasSubset,,,Anchoring Bias,FALSE,0.38,"The influence of a reference point or anchor on decisions, leading to insufficient adjustment from that anchor point.",,A cognitive bias characterized by the influence of a reference point or anchor on decisions leading to insufficient adjustment from that anchor point.,An individual bias characterized by the influence of a reference point or anchor on decisions leading to insufficient adjustment from that anchor point.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:ConfirmationBias,AIO:BiasSubset,,,Confirmation Bias,FALSE,0.42,"The tendency to prefer information that confirms existing beliefs, influencing the search for, interpretation of, and recall of information.","The tendency to prefer information that confirms existing beliefs, influencing the search for, interpretation of, and recall of information.",A cognitive bias characterized by the tendency to prefer information that confirms existing beliefs influencing the search for interpretation of and recall of information.,An individual bias characterized by the tendency to prefer information that confirms existing beliefs influencing the search for interpretation of and recall of information.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:SelectiveAdherenceBias,AIO:BiasSubset,,,Selective Adherence Bias,FALSE,0.37,The tendency to selectively adopt algorithmic advice that matches pre-existing beliefs and stereotypes.,,An individual bias characterized by the tendency to selectively adopt algorithmic advice that matches pre-existing beliefs and stereotypes.,An individual bias characterized by the tendency to selectively adopt algorithmic advice that matches pre-existing beliefs and stereotypes.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:ImplicitBias,AIO:BiasSubset,Confirmatory Bias,,Implicit Bias,FALSE,0.73,"Unconscious beliefs, attitudes, feelings, associations, or stereotypes that affect information processing, decision-making, and actions.","Unconscious beliefs, attitudes, feelings, associations, or stereotypes that affect information processing, decision-making, and actions.",An individual bias characterized by unconscious beliefs attitudes feelings associations or stereotypes that affect information processing decision-making and actions.,An individual bias characterized by unconscious beliefs attitudes feelings associations or stereotypes that affect information processing decision-making and actions.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:LossOfSituationalAwarenessBias,AIO:BiasSubset,,,Loss Of Situational Awareness Bias,FALSE,0.33,"When automation leads to humans being unaware of their situation, making them unprepared to assume control in cooperative systems.",,An individual bias occurring when automation leads to humans being unaware of their situation making them unprepared to assume control in cooperative systems.,An individual bias occurring when automation leads to humans being unaware of their situation making them unprepared to assume control in cooperative systems.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:ModeConfusionBias,AIO:BiasSubset,,,Mode Confusion Bias,FALSE,0.30,"When modal interfaces confuse human operators, causing actions appropriate for a different mode but incorrect for the current situation.",,A bias occurring when modal interfaces confuse human operators causing actions appropriate for a different mode but incorrect for the current situation.,An individual bias occurring when modal interfaces confuse human operators causing actions appropriate for a different mode but incorrect for the current situation.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:AnnotatorReportingBias,AIO:BiasSubset,,,Annotator Reporting Bias,FALSE,0.29,When users rely on automation as a heuristic replacement for their own information seeking and processing.,,An individual bias occurring when users rely on automation as a heuristic replacement for their own information seeking and processing.,An individual bias occurring when users rely on automation as a heuristic replacement for their own information seeking and processing.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:HumanReportingBias,AIO:BiasSubset,,,Human Reporting Bias,FALSE,0.29,When users rely on automation as a heuristic replacement for their own information seeking and processing.,,An individual bias that arises when users depend on automated systems as heuristic substitutes for their own information-seeking and processing efforts.,An individual bias that arises when users depend on automated systems as heuristic substitutes for their own information-seeking and processing efforts.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:StreetlightEffectBias,AIO:BiasSubset,Streetlight Effect,,Streetlight Effect Bias,FALSE,0.31,Bias where people search only where it is easiest to look.,,An individual bias where people search only where it is easiest to look.,An individual bias where people search only where it is easiest to look.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:InterpretationBias,AIO:BiasSubset,,,Interpretation Bias,FALSE,0.35,A form of information processing bias where users interpret algorithmic outputs according to their internalized biases and views.,,An individual bias where users interpret algorithmic outputs according to their internalized biases and views.,An individual bias where users interpret algorithmic outputs according to their internalized biases and views.,https://doi.org/10.6028/NIST.SP.1270,,,AIO:IndividualBias +AIO:NoisyInputLayer,AIO:LayerSubset,,,Noisy Input Layer,FALSE,#N/A,,,An input layer that adds noise to each value.,An input layer that adds noise to each value.,https://doi.org/10.1109/21.155944,,,AIO:InputLayer +AIO:BackfedInputLayer,AIO:LayerSubset,,,Backfed Input Layer,FALSE,#N/A,,,An input layer that receives values from another layer.,An input layer that receives values from another layer.,,,,AIO:InputLayer +AIO:MatchedInputOutputLayer,AIO:LayerSubset,,,Matched Input-Output Layer,FALSE,#N/A,,,An input layer with a shape corresponding to that of the output layer.,An input layer with a shape corresponding to that of the output layer.,,,,AIO:InputLayer +AIO:KnearestNeighborRegressionAlgorithm,AIO:MachineLearningSubset,KNN Regression|K-NN Regression,,K-nearest Neighbor Regression Algorithm,FALSE,#N/A,,,An regression analysis that assigns the average of the values of k nearest neighbors to objects.,An regression analysis that assigns the average of the values of k nearest neighbors to objects.,https://en.wikipedia.org/wiki/K-nearest_neighbors_algorithm|,,,AIO:RegressionAnalysis +AIO:DimensionalityReduction,AIO:MachineLearningSubset,Dimension Reduction,,Dimensionality Reduction,FALSE,0.29,The process of transforming data from a high-dimensional space into a lower-dimensional space while retaining meaningful properties of the original data.,,A machine learning task focused on the process of transforming data from a high-dimensional space into a lower-dimensional space while retaining meaningful properties of the original data.,An unsupervised learning focused on the process of transforming data from a high-dimensional space into a lower-dimensional space while retaining meaningful properties of the original data.,https://en.wikipedia.org/wiki/Dimensionality_reduction|,,,AIO:UnsupervisedLearning +AIO:DeepBeliefNetwork,AIO:NetworkSubset,DBN,,Deep Belief Network,FALSE,0.57,"A deep belief network (DBN) is a generative graphical model composed of multiple layers of latent variables, learning to probabilistically reconstruct inputs and perform classification.",,An unsupervised pretrained network composed of multiple layers of latent variables that learns to probabilistically reconstruct inputs and perform classification.,An unsupervised pretrained network composed of multiple layers of latent variables that learns to probabilistically reconstruct inputs and perform classification.,https://en.wikipedia.org/wiki/Deep_belief_network,"Layers: Backfed Input, Probabilistic Hidden, Hidden, Matched Output-Input",AIO:BackfedInputLayer|AIO:ProbabilisticHiddenLayer|AIO:HiddenLayer|AIO:MatchedInputOutputLayer,AIO:UnsupervisedPretrainedNetwork +AIO:GenerativeAdversarialNetwork,AIO:NetworkSubset,GAN,,Generative Adversarial Network,FALSE,0.32,A generative adversarial network (GAN) is a machine learning framework where two neural networks contest in a game to generate new data with the same statistics as the training set.,,An unsupervised pretrained network framework where two neural networks contest in a game to generate new data with the same statistics as the training set.,An unsupervised pretrained network framework where two neural networks contest in a game to generate new data with the same statistics as the training set.,https://en.wikipedia.org/wiki/Generative_adversarial_network,"Layers: Backfed Input, Hidden, Matched Output-Input, Hidden, Matched Output-Input",AIO:BackfedInputLayer|AIO:HiddenLayer|AIO:MatchedInputOutputLayer,AIO:UnsupervisedPretrainedNetwork +AIO:AutoEncoderNetwork,AIO:NetworkSubset,AE,,Auto Encoder Network,FALSE,0.48,"An autoencoder is an artificial neural network used for learning efficient codings of unlabeled data, training the network to ignore insignificant data and regenerate input from encoding.",,An unsupervised pretrained network that learns efficient codings of unlabeled data by training to ignore insignificant data and regenerate input from encoding.,An unsupervised pretrained network that learns efficient codings of unlabeled data by training to ignore insignificant data and regenerate input from encoding.,https://en.wikipedia.org/wiki/Autoencoder,"Layers: Input, Hidden, Matched Output-Input",AIO:InputLayer|AIO:HiddenLayer|AIO:MatchedInputOutputLayer,AIO:UnsupervisedPretrainedNetwork +AIO:RNNLayer,AIO:LayerSubset,,,RNN Layer,FALSE,0.43,Base class for recurrent layers.,Base class for recurrent layers.,The base class for recurrent layers.,The base class for recurrent layers.,https://www.tensorflow.org/api_docs/python/tf/keras/layers/RNN,,,AIO:RecurrentLayer +AIO:Preprocessing,AIO:PreprocessingSubset,,,Preprocessing,FALSE,0.26,"The series of steps applied to raw data before it is used in a machine learning model, including tasks such as normalization, scaling, encoding, and transformation, to ensure the data is in an appropriate format and quality for analysis.",,The series of steps applied to raw data before it is used in a machine learning model including tasks such as normalization scaling encoding and transformation to ensure the data is in an appropriate format and quality for analysis.,The series of steps applied to raw data before it is used in a machine learning model including tasks such as normalization scaling encoding and transformation to ensure the data is in an appropriate format and quality for analysis.,https://doi.org/10.1109/ICDE.2019.00245,,,owl:Thing \ No newline at end of file diff --git a/src/ontology/aio.Makefile b/src/ontology/aio.Makefile index 8623e0c..e390a3e 100644 --- a/src/ontology/aio.Makefile +++ b/src/ontology/aio.Makefile @@ -38,3 +38,35 @@ components/%.owl: components/%.csv -t $< \ annotate --annotation-file aio-annotations.ttl \ -o $@ + +clean-extras: + rm -rf aio.db aio-root-statistics* aio-2024-06-26.owl current-vs-2024-06-26-diff* + +remove-old-input: + rm -rf aio-src.csv + rm -rf components/aio-component.owl + +components-from-new-input: remove-old-input + make aio-src.csv + make components/aio-component.owl + +aio.db: aio.owl # requires that rdftab and relation0graph have been installed locally and are on the default path + poetry run semsql make $@ + rm -rf aio-relation-graph.tsv.gz + +aio-root-statistics.tsv: aio.db # could use sqlite:obo:aio if the file in the BBOP S3 bucket is new enough + poetry run runoak \ + --input $< statistics \ + -O tsv -o $@ $$(poetry run runoak --input $< roots | cut -f 1 | sed '1d' | tr '\n' ' ') + +aio-root-statistics-transposed-filtered.tsv: aio-root-statistics.tsv + python ../scripts/transpose_filter_statistics.py < $< > $@ + +aio-2024-06-26.owl: + wget -O $@ "https://raw.githubusercontent.com/berkeleybop/artificial-intelligence-ontology/v2024-06-26/aio.owl" + +current-vs-2024-06-26-diff.txt: aio.owl aio-2024-06-26.owl + robot diff --left $< --right aio-2024-06-26.owl --output $@ + +current-vs-2024-06-26-diff-skip-defs.txt: current-vs-2024-06-26-diff.txt + cat $< | grep -v IAO_0000115 > $@ \ No newline at end of file diff --git a/src/ontology/components/aio-component.owl b/src/ontology/components/aio-component.owl index a46c245..872efff 100644 --- a/src/ontology/components/aio-component.owl +++ b/src/ontology/components/aio-component.owl @@ -1,7 +1,6 @@ - This ontology models classes and relationships describing deep learning networks, their component layers and activation functions, as well as potential biases. - Artificial Intelligence Ontology + This ontology models classes and relationships describing deep learning networks, their component layers and activation functions, as well as potential biases. + Artificial Intelligence Ontology @@ -35,21 +34,21 @@ - + - + - + - + - + - + @@ -77,6 +76,23 @@ + + + + + + + + + + + - - A type of machine learning focused on methods that interactively query a user or another information source to label new data points with the desired outputs. + + A machine learning task focused on methods that interactively query a user or another information source to label new data points with the desired outputs. Query Learning Active Learning @@ -134,8 +151,8 @@ - A type of machine learning focused on methods that interactively query a user or another information source to label new data points with the desired outputs. - https://en.wikipedia.org/wiki/Active_learning_(machine_learning) + A machine learning task focused on methods that interactively query a user or another information source to label new data points with the desired outputs. + @@ -152,8 +169,7 @@ A use and interpretation bias occurring when systems/platforms get training data from their most active users rather than less active or inactive users. - GTP-4o with Seppala et al. 2017 - https://en.wikipedia.org/wiki/Interpretive_bias + @@ -170,7 +186,7 @@ A regularization layer that applies an update to the cost function based on input activity. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/ActivityRegularization + @@ -181,7 +197,6 @@ A pooling layer that applies a 1D adaptive average pooling over an input signal composed of several input planes. AdaptiveAvgPool1D - AdaptiveAvgPool1d AdaptiveAvgPool1D Layer @@ -189,7 +204,7 @@ A pooling layer that applies a 1D adaptive average pooling over an input signal composed of several input planes. - https://pytorch.org/docs/stable/nn.html#pooling-layers + @@ -200,7 +215,6 @@ A pooling layer that applies a 2D adaptive average pooling over an input signal composed of several input planes. AdaptiveAvgPool2D - AdaptiveAvgPool2d AdaptiveAvgPool2D Layer @@ -208,7 +222,7 @@ A pooling layer that applies a 2D adaptive average pooling over an input signal composed of several input planes. - https://pytorch.org/docs/stable/nn.html#pooling-layers + @@ -219,7 +233,6 @@ A pooling layer that applies a 3D adaptive average pooling over an input signal composed of several input planes. AdaptiveAvgPool3D - AdaptiveAvgPool3d AdaptiveAvgPool3D Layer @@ -227,7 +240,7 @@ A pooling layer that applies a 3D adaptive average pooling over an input signal composed of several input planes. - https://pytorch.org/docs/stable/nn.html#pooling-layers + @@ -238,7 +251,6 @@ A pooling layer that applies a 1D adaptive max pooling over an input signal composed of several input planes. AdaptiveMaxPool1D - AdaptiveMaxPool1d AdaptiveMaxPool1D Layer @@ -246,7 +258,7 @@ A pooling layer that applies a 1D adaptive max pooling over an input signal composed of several input planes. - https://pytorch.org/docs/stable/nn.html#pooling-layers + @@ -257,7 +269,6 @@ A pooling layer that applies a 2D adaptive max pooling over an input signal composed of several input planes. AdaptiveMaxPool2D - AdaptiveMaxPool2d AdaptiveMaxPool2D Layer @@ -265,7 +276,7 @@ A pooling layer that applies a 2D adaptive max pooling over an input signal composed of several input planes. - https://pytorch.org/docs/stable/nn.html#pooling-layers + @@ -276,7 +287,6 @@ A pooling layer that applies a 3D adaptive max pooling over an input signal composed of several input planes. AdaptiveMaxPool3D - AdaptiveMaxPool3d AdaptiveMaxPool3D Layer @@ -284,7 +294,7 @@ A pooling layer that applies a 3D adaptive max pooling over an input signal composed of several input planes. - https://pytorch.org/docs/stable/nn.html#pooling-layers + @@ -295,13 +305,14 @@ A merging layer that adds a list of inputs taking as input a list of tensors all of the same shape. + Layer that adds a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape). Add Layer A merging layer that adds a list of inputs taking as input a list of tensors all of the same shape. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Add + @@ -310,9 +321,6 @@ - A layer that adds inputs from one or more other layers to cells or neurons of a target layer. - - Addition Layer @@ -323,13 +331,14 @@ An attention layer that implements additive attention also known as Bahdanau-style attention. + Additive attention layer, a.k.a. Bahdanau-style attention. AdditiveAttention Layer An attention layer that implements additive attention also known as Bahdanau-style attention. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/AdditiveAttention + @@ -340,13 +349,14 @@ A regularization layer that applies Alpha Dropout to the input keeping mean and variance of inputs to ensure self-normalizing property. + Applies Alpha Dropout to the input. Alpha Dropout is a Dropout that keeps mean and variance of inputs to their original values, in order to ensure the self-normalizing property even after this dropout. Alpha Dropout fits well to Scaled Exponential Linear Units by randomly setting activations to the negative saturation value. AlphaDropout Layer A regularization layer that applies Alpha Dropout to the input keeping mean and variance of inputs to ensure self-normalizing property. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/AlphaDropout + @@ -363,8 +373,7 @@ A processing bias arising when the distribution over prediction outputs is skewed compared to the prior distribution of the prediction target. - GTP-4o with Seppala et al. 2017 - https://royalsocietypublishing.org/doi/10.1098/rspb.2019.0165#d1e5237 + @@ -373,16 +382,15 @@ - A cognitive bias characterized by the influence of a reference point or anchor on decisions leading to insufficient adjustment from that anchor point. + An individual bias characterized by the influence of a reference point or anchor on decisions leading to insufficient adjustment from that anchor point. Anchoring Bias - A cognitive bias characterized by the influence of a reference point or anchor on decisions leading to insufficient adjustment from that anchor point. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + An individual bias characterized by the influence of a reference point or anchor on decisions leading to insufficient adjustment from that anchor point. + @@ -399,8 +407,7 @@ An individual bias occurring when users rely on automation as a heuristic replacement for their own information seeking and processing. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -413,13 +420,14 @@ ANN NN + An artificial neural network (ANN) is based on a collection of connected units or nodes called artificial neurons, modeled after biological neurons, with connections transmitting signals processed by non-linear functions. Artificial Neural Network A network based on a collection of connected units called artificial neurons modeled after biological neurons. - https://en.wikipedia.org/wiki/Artificial_neural_network + @@ -428,15 +436,15 @@ - A supervised learning method focused on a rule-based approach for discovering interesting relations between variables in large databases. + A supervised learning focused on a rule-based approach for discovering interesting relations between variables in large databases. Association Rule Learning - A supervised learning method focused on a rule-based approach for discovering interesting relations between variables in large databases. - https://en.wikipedia.org/wiki/Association_rule_learning + A supervised learning focused on a rule-based approach for discovering interesting relations between variables in large databases. + @@ -447,13 +455,14 @@ A layer that implements dot-product attention also known as Luong-style attention. + Dot-product attention layer, a.k.a. Luong-style attention. Attention Layer A layer that implements dot-product attention also known as Luong-style attention. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Attention + @@ -462,6 +471,24 @@ + + + + + + + + + + + + + + + + + + An unsupervised pretrained network that learns efficient codings of unlabeled data by training to ignore insignificant data and regenerate input from encoding. AE @@ -472,7 +499,7 @@ An unsupervised pretrained network that learns efficient codings of unlabeled data by training to ignore insignificant data and regenerate input from encoding. - https://en.wikipedia.org/wiki/Autoencoder + @@ -481,17 +508,17 @@ - A bias characterized by over-reliance on automated systems leading to attenuated human skills. + An individual bias characterized by over-reliance on automated systems leading to attenuated human skills. Automation Complaceny + Over-reliance on automated systems, leading to attenuated human skills, such as with spelling and autocorrect. Automation Complacency Bias - A bias characterized by over-reliance on automated systems leading to attenuated human skills. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + An individual bias characterized by over-reliance on automated systems leading to attenuated human skills. + @@ -537,7 +564,6 @@ A language model that generates text sequentially predicting one token at a time based on the previously generated tokens excelling at natural language generation tasks by modeling the probability distribution over sequences of tokens. - Autoregressive Language Model generative language model sequence-to-sequence model @@ -562,7 +588,7 @@ - A cognitive bias characterized by a mental shortcut where easily recalled information is overweighted in judgment and decision-making. + An individual bias characterized by a mental shortcut where easily recalled information is overweighted in judgment and decision-making. Availability Bias Availability Heuristic @@ -571,9 +597,8 @@ - A cognitive bias characterized by a mental shortcut where easily recalled information is overweighted in judgment and decision-making. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + An individual bias characterized by a mental shortcut where easily recalled information is overweighted in judgment and decision-making. + @@ -584,13 +609,14 @@ A merging layer that averages a list of inputs element-wise taking as input a list of tensors all of the same shape. + Layer that averages a list of inputs element-wise. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape). Average Layer A merging layer that averages a list of inputs element-wise taking as input a list of tensors all of the same shape. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Average + @@ -601,15 +627,15 @@ A pooling layer that performs average pooling for temporal data. AvgPool1D - AvgPool1d + Average pooling for temporal data. Downsamples the input representation by taking the average value over the window defined by pool_size. The window is shifted by strides. The resulting output when using "valid" padding option has a shape of: output_shape = (input_shape - pool_size + 1) / strides). The resulting output shape when using the "same" padding option is: output_shape = input_shape / strides. AveragePooling1D Layer A pooling layer that performs average pooling for temporal data. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/AveragePooling1D + @@ -620,15 +646,15 @@ A pooling layer that performs average pooling for spatial data. AvgPool2D - AvgPool2d + Average pooling operation for spatial data. Downsamples the input along its spatial dimensions (height and width) by taking the average value over an input window (of size defined by pool_size) for each channel of the input. The window is shifted by strides along each dimension. The resulting output when using "valid" padding option has a shape (number of rows or columns) of: output_shape = math.floor((input_shape - pool_size) / strides) + 1 (when input_shape >= pool_size). The resulting output shape when using the "same" padding option is: output_shape = math.floor((input_shape - 1) / strides) + 1. AveragePooling2D Layer A pooling layer that performs average pooling for spatial data. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/AveragePooling2D + @@ -640,13 +666,14 @@ A pooling layer that performs average pooling for 3D data (spatial or spatio-temporal). AvgPool3D + Average pooling operation for 3D data (spatial or spatio-temporal). Downsamples the input along its spatial dimensions (depth, height, and width) by taking the average value over an input window (of size defined by pool_size) for each channel of the input. The window is shifted by strides along each dimension. AveragePooling3D Layer A pooling layer that performs average pooling for 3D data (spatial or spatio-temporal). - https://www.tensorflow.org/api_docs/python/tf/keras/layers/AveragePooling3D + @@ -657,7 +684,6 @@ A pooling layer that applies a 1D average pooling over an input signal composed of several input planes. AvgPool1D - AvgPool1d AvgPool1D Layer @@ -665,7 +691,7 @@ A pooling layer that applies a 1D average pooling over an input signal composed of several input planes. - https://pytorch.org/docs/stable/nn.html#pooling-layers + @@ -676,7 +702,6 @@ A pooling layer that applies a 2D average pooling over an input signal composed of several input planes. AvgPool2D - AvgPool2d AvgPool2D Layer @@ -684,7 +709,7 @@ A pooling layer that applies a 2D average pooling over an input signal composed of several input planes. - https://pytorch.org/docs/stable/nn.html#pooling-layers + @@ -695,7 +720,6 @@ A pooling layer that applies a 3D average pooling over an input signal composed of several input planes. AvgPool3D - AvgPool3d AvgPool3D Layer @@ -703,7 +727,7 @@ A pooling layer that applies a 3D average pooling over an input signal composed of several input planes. - https://pytorch.org/docs/stable/nn.html#pooling-layers + @@ -712,9 +736,6 @@ - An input layer that receives values from another layer. - - Backfed Input Layer @@ -725,15 +746,15 @@ A batch normalization layer that applies Batch Normalization over a 2D or 3D input. BatchNorm1D - BatchNorm1d + Applies Batch Normalization over a 2D or 3D input as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift . BatchNorm1D Layer A batch normalization layer that applies Batch Normalization over a 2D or 3D input. - https://pytorch.org/docs/stable/nn.html#normalization-layers + @@ -744,15 +765,15 @@ A batch normalization layer that applies Batch Normalization over a 4D input. BatchNorm2D - BatchNorm2d + Applies Batch Normalization over a 4D input (a mini-batch of 2D inputs with additional channel dimension) as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift . BatchNorm2D Layer A batch normalization layer that applies Batch Normalization over a 4D input. - https://pytorch.org/docs/stable/nn.html#normalization-layers + @@ -763,15 +784,15 @@ A batch normalization layer that applies Batch Normalization over a 5D input. BatchNorm3D - BatchNorm3d + Applies Batch Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension) as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift . BatchNorm3D Layer A batch normalization layer that applies Batch Normalization over a 5D input. - https://pytorch.org/docs/stable/nn.html#normalization-layers + @@ -780,16 +801,7 @@ - A normalization layer that normalizes its inputs applying a transformation that maintains the mean close to 0 and the standard deviation close to 1. - - BatchNormalization Layer - - - - A normalization layer that normalizes its inputs applying a transformation that maintains the mean close to 0 and the standard deviation close to 1. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/BatchNormalization - @@ -805,7 +817,7 @@ A network that is a probabilistic graphical model representing variables and their conditional dependencies via a directed acyclic graph. - https://en.wikipedia.org/wiki/Bayesian_network + @@ -816,14 +828,14 @@ An individual bias characterized by systematic distortions in user behavior across platforms or contexts or across users represented in different datasets. + Systematic distortions in user behavior across platforms or contexts, or across users represented in different datasets. Behavioral Bias An individual bias characterized by systematic distortions in user behavior across platforms or contexts or across users represented in different datasets. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -840,8 +852,7 @@ A systematic error introduced into sampling or testing by selecting or encouraging one outcome or answer over others. - GTP-4o with Seppala et al. 2017 - https://www.merriam-webster.com/dictionary/bias + @@ -849,7 +860,7 @@ - + A machine learning task focused on methods that simultaneously cluster the rows and columns of a matrix to identify submatrices with coherent patterns. Block Clustering Co-clustering @@ -863,7 +874,7 @@ A machine learning task focused on methods that simultaneously cluster the rows and columns of a matrix to identify submatrices with coherent patterns. - https://en.wikipedia.org/wiki/Biclustering + @@ -874,13 +885,14 @@ A recurrent layer that is a bidirectional wrapper for RNNs. + Bidirectional wrapper for RNNs. Bidirectional Layer A recurrent layer that is a bidirectional wrapper for RNNs. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Bidirectional + @@ -899,8 +911,8 @@ A transformer language model such as BERT that uses the transformer architecture to build deep bidirectional representations by predicting masked tokens based on their context. - https://arxiv.org/abs/1810.04805 - https://en.wikipedia.org/wiki/BERT_(language_model) + + @@ -909,15 +921,15 @@ - A machine learning task focused on methods that classify elements into two groups based on a classification rule. + A classification focused on methods that classify elements into two groups based on a classification rule. Binary Classification - A machine learning task focused on methods that classify elements into two groups based on a classification rule. - https://en.wikipedia.org/wiki/Binary_classification + A classification focused on methods that classify elements into two groups based on a classification rule. + @@ -926,6 +938,18 @@ + + + + + + + + + + + + A symmetrically connected network that is a type of stochastic recurrent neural network and Markov random field. BM Sherrington–Kirkpatrick model with external field @@ -939,7 +963,7 @@ A symmetrically connected network that is a type of stochastic recurrent neural network and Markov random field. - https://en.wikipedia.org/wiki/Boltzmann_machine + @@ -956,7 +980,7 @@ A layer that performs categorical data preprocessing operations. - https://keras.io/guides/preprocessing_layers/ + @@ -967,13 +991,14 @@ A categorical features preprocessing layer that encodes integer features providing options for condensing data into a categorical encoding. + A preprocessing layer which encodes integer features. This layer provides options for condensing data into a categorical encoding when the total number of tokens are known in advance. It accepts integer values as inputs, and it outputs a dense or sparse representation of those inputs. For integer inputs where the total number of tokens is not known, use tf.keras.layers.IntegerLookup instead. CategoryEncoding Layer A categorical features preprocessing layer that encodes integer features providing options for condensing data into a categorical encoding. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/CategoryEncoding + @@ -995,7 +1020,7 @@ A probabilistic graphical model used to encode assumptions about the data-generating process. - https://en.wikipedia.org/wiki/Causal_graph + @@ -1020,13 +1045,14 @@ An image preprocessing layer that crops the central portion of images to a target size. + A preprocessing layer which crops images. This layers crops the central portion of the images to a target size. If an image is smaller than the target size, it will be resized and cropped so as to return the largest possible window in the image that matches the target aspect ratio. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. CenterCrop Layer An image preprocessing layer that crops the central portion of images to a target size. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/CenterCrop + @@ -1035,15 +1061,16 @@ - A supervised learning task focused on methods that distinguish and distribute kinds of "things" into different groups. + A supervised learning focused on methods that distinguish and distribute kinds of "things" into different groups. + Methods that distinguish and distribute kinds of "things" into different groups. Classification - A supervised learning task focused on methods that distinguish and distribute kinds of "things" into different groups. - https://en.wikipedia.org/wiki/Classification_(general_theory) + A supervised learning focused on methods that distinguish and distribute kinds of "things" into different groups. + @@ -1052,7 +1079,7 @@ - The process of removing noise inconsistencies and irrelevant information from data to enhance its quality and prepare it for analysis or further processing. + A data preparation that removes noise inconsistencies and irrelevant information from data to enhance its quality and prepare it for analysis or further processing. Data Cleansing Standardization Data cleaning @@ -1066,7 +1093,7 @@ - + A machine learning task focused on methods that group a set of objects such that objects in the same group are more similar to each other than to those in other groups. Cluster analysis @@ -1076,7 +1103,7 @@ A machine learning task focused on methods that group a set of objects such that objects in the same group are more similar to each other than to those in other groups. - https://en.wikipedia.org/wiki/Cluster_analysis + @@ -1085,16 +1112,15 @@ - A systematic deviation from rational judgment and decision-making including adaptive mental shortcuts known as heuristics. + An individual bias characterized by deviations from rational judgment and decision-making including adaptive mental shortcuts known as heuristics. Cognitive Bias - A systematic deviation from rational judgment and decision-making including adaptive mental shortcuts known as heuristics. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + An individual bias characterized by deviations from rational judgment and decision-making including adaptive mental shortcuts known as heuristics. + @@ -1126,8 +1152,7 @@ A bias caused by differences between results and facts in the process of data analysis (including the source of data the estimator chose) and analysis methods. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -1138,13 +1163,14 @@ A merging layer that concatenates a list of inputs taking as input a list of tensors all of the same shape except for the concatenation axis. + Layer that concatenates a list of inputs. It takes as input a list of tensors, all of the same shape except for the concatenation axis, and returns a single tensor that is the concatenation of all inputs. Concatenate Layer A merging layer that concatenates a list of inputs taking as input a list of tensors all of the same shape except for the concatenation axis. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Concatenate + @@ -1162,8 +1188,7 @@ A use and interpretation bias due to the use of a system outside its planned domain of application causing performance gaps between laboratory settings and the real world. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -1172,16 +1197,16 @@ - A cognitive bias characterized by the tendency to prefer information that confirms existing beliefs influencing the search for interpretation of and recall of information. + An individual bias characterized by the tendency to prefer information that confirms existing beliefs influencing the search for interpretation of and recall of information. + The tendency to prefer information that confirms existing beliefs, influencing the search for, interpretation of, and recall of information. Confirmation Bias - A cognitive bias characterized by the tendency to prefer information that confirms existing beliefs influencing the search for interpretation of and recall of information. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + An individual bias characterized by the tendency to prefer information that confirms existing beliefs influencing the search for interpretation of and recall of information. + @@ -1190,16 +1215,15 @@ - A bias arising when an algorithm or platform provides users a venue to express their biases occurring from either side in a digital interaction. + An individual bias arising when an algorithm or platform provides users a venue to express their biases occurring from either side in a digital interaction. Consumer Bias - A bias arising when an algorithm or platform provides users a venue to express their biases occurring from either side in a digital interaction. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + An individual bias arising when an algorithm or platform provides users a venue to express their biases occurring from either side in a digital interaction. + @@ -1210,14 +1234,14 @@ A use and interpretation bias arising from structural lexical semantic and syntactic differences in user-generated content. + Bias from structural, lexical, semantic, and syntactic differences in user-generated content. Content Production Bias A use and interpretation bias arising from structural lexical semantic and syntactic differences in user-generated content. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -1230,13 +1254,14 @@ Incremental Learning Life-Long Learning + Learning a model for sequential tasks without forgetting knowledge from preceding tasks, with no access to old task data during new task training. Continual Learning A deep neural network that learns sequential tasks without forgetting knowledge from preceding tasks and without access to old task data during new task training. - https://paperswithcode.com/task/continual-learning + @@ -1262,13 +1287,14 @@ A deep neural network self-supervised learning approach that learns to distinguish between similar and dissimilar data samples. + Contrastive learning is a self-supervised learning approach in which the model learns to distinguish between similar and dissimilar pairs of data samples. By maximizing the similarity between positive pairs (similar samples) and minimizing the similarity between negative pairs (dissimilar samples), the model learns to capture meaningful representations of the data. This method is particularly effective for representation learning and is widely used in tasks such as image classification, clustering, and retrieval. Contrastive learning techniques often employ loss functions such as the contrastive loss or the triplet loss to achieve these objectives. Contrastive Learning A deep neural network self-supervised learning approach that learns to distinguish between similar and dissimilar data samples. - https://arxiv.org/abs/2202.14037 + @@ -1278,7 +1304,6 @@ A large language model that is trained to pull semantically similar samples closer together and push dissimilar samples apart in the representation space learning high-quality features useful for downstream tasks. - Contrastive Learning LLM Representation learning Contrastive Learning LLM @@ -1295,6 +1320,7 @@ conditional generation guided generation + A controllable LLM allows for explicit control over certain attributes of the generated text, such as style, tone, topic, or other desired characteristics, through conditioning or specialized training objectives. Controllable LLM @@ -1306,13 +1332,14 @@ A convolutional layer that implements a 1D Convolutional LSTM similar to an LSTM but with convolutional input and recurrent transformations. + 1D Convolutional LSTM. Similar to an LSTM layer, but the input transformations and recurrent transformations are both convolutional. ConvLSTM1D Layer A convolutional layer that implements a 1D Convolutional LSTM similar to an LSTM but with convolutional input and recurrent transformations. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/ConvLSTM1D + @@ -1323,13 +1350,14 @@ A convolutional layer that implements a 2D Convolutional LSTM similar to an LSTM but with convolutional input and recurrent transformations. + 2D Convolutional LSTM. Similar to an LSTM layer, but the input transformations and recurrent transformations are both convolutional. ConvLSTM2D Layer A convolutional layer that implements a 2D Convolutional LSTM similar to an LSTM but with convolutional input and recurrent transformations. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/ConvLSTM2D + @@ -1340,13 +1368,14 @@ A convolutional layer that implements a 3D Convolutional LSTM similar to an LSTM but with convolutional input and recurrent transformations. + 3D Convolutional LSTM. Similar to an LSTM layer, but the input transformations and recurrent transformations are both convolutional. ConvLSTM3D Layer A convolutional layer that implements a 3D Convolutional LSTM similar to an LSTM but with convolutional input and recurrent transformations. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/ConvLSTM3D + @@ -1355,20 +1384,19 @@ - A layer that implements 1D convolution (e.g. temporal convolution). + A convolutional layer that implements 1D convolution (e.g. temporal convolution). + Conv1D Conv1D Layer - Conv1d Convolution1D - Convolution1d - nn.Conv1d + nn.Conv1D Convolution1D Layer - A layer that implements 1D convolution (e.g. temporal convolution). - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv1D + A convolutional layer that implements 1D convolution (e.g. temporal convolution). + @@ -1377,20 +1405,20 @@ - A layer that implements transposed 1D convolution sometimes called deconvolution. + A convolutional layer that implements transposed 1D convolution sometimes called deconvolution. Conv1DTranspose Layer - ConvTranspose1d + ConvTranspose1D Convolution1DTranspose - Convolution1dTranspose - nn.ConvTranspose1d + nn.ConvTranspose1D + Transposed convolution layer (sometimes called Deconvolution). The need for transposed convolutions generally arises from the desire to use a transformation going in the opposite direction of a normal convolution, i.e., from something that has the shape of the output of some convolution to something that has the shape of its input while maintaining a connectivity pattern that is compatible with said convolution. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 3) for data with 128 time steps and 3 channels. Convolution1DTranspose Layer - A layer that implements transposed 1D convolution sometimes called deconvolution. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv1DTranspose + A convolutional layer that implements transposed 1D convolution sometimes called deconvolution. + @@ -1399,20 +1427,20 @@ - A layer that implements 2D convolution (e.g. spatial convolution over images). + A convolutional layer that implements 2D convolution (e.g. spatial convolution over images). + Conv2D Conv2D Layer - Conv2d Convolution2D - Convolution2d - nn.Conv2d + nn.Conv2D + 2D convolution layer (e.g. spatial convolution over images). This layer creates a convolution kernel that is convolved with the layer input to produce a tensor of outputs. If use_bias is True, a bias vector is created and added to the outputs. Finally, if activation is not None, it is applied to the outputs as well. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 128, 3) for 128x128 RGB pictures in data_format="channels_last". You can use None when a dimension has variable size. Convolution2D Layer - A layer that implements 2D convolution (e.g. spatial convolution over images). - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv2D + A convolutional layer that implements 2D convolution (e.g. spatial convolution over images). + @@ -1421,20 +1449,20 @@ - A layer that implements transposed 2D convolution + A convolutional layer that implements transposed 2D convolution Conv2DTranspose Layer - ConvTranspose2d + ConvTranspose2D Convolution2DTranspose - Convolution2dTranspose - nn.ConvTranspose2d + nn.ConvTranspose2D + Transposed convolution layer (sometimes called Deconvolution). Convolution2DTranspose Layer - A layer that implements transposed 2D convolution - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv2DTranspose + A convolutional layer that implements transposed 2D convolution + @@ -1443,20 +1471,20 @@ - A layer that implements 3D convolution (e.g. spatial convolution over volumes). + A convolutional layer that implements 3D convolution (e.g. spatial convolution over volumes). + Conv3D Conv3D Layer - Conv3d Convolution3D - Convolution3d - nn.Conv3d + nn.Conv3D + 3D convolution layer (e.g. spatial convolution over volumes). This layer creates a convolution kernel that is convolved with the layer input to produce a tensor of outputs. If use_bias is True, a bias vector is created and added to the outputs. Finally, if activation is not None, it is applied to the outputs as well. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 128, 128, 1) for 128x128x128 volumes with a single channel, in data_format="channels_last". Convolution3D Layer - A layer that implements 3D convolution (e.g. spatial convolution over volumes). - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv3D + A convolutional layer that implements 3D convolution (e.g. spatial convolution over volumes). + @@ -1465,20 +1493,20 @@ - A layer that implements transposed 3D convolution + A convolutional layer that implements transposed 3D convolution Conv3DTranspose Layer - ConvTranspose3d + ConvTranspose3D Convolution3DTranspose - Convolution3dTranspose - nn.ConvTranspose3d + nn.ConvTranspose3D + Transposed convolution layer (sometimes called Deconvolution). The need for transposed convolutions generally arises from the desire to use a transformation going in the opposite direction of a normal convolution, i.e., from something that has the shape of the output of some convolution to something that has the shape of its input while maintaining a connectivity pattern that is compatible with said convolution. When using this layer as the first layer in a model, provide the keyword argument input_shape (tuple of integers or None, does not include the sample axis), e.g. input_shape=(128, 128, 128, 3) for a 128x128x128 volume with 3 channels if data_format="channels_last". Convolution3DTranspose Layer - A layer that implements transposed 3D convolution - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv3DTranspose + A convolutional layer that implements transposed 3D convolution + @@ -1487,16 +1515,7 @@ - A layer that contains a set of filters (or kernels) parameters of which are to be learned throughout the training. - - Convolutional Layer - - - - A layer that contains a set of filters (or kernels) parameters of which are to be learned throughout the training. - https://www.sciencedirect.com/topics/engineering/convolutional-layer#:~:text=A%20convolutional%20layer%20is%20the,and%20creates%20an%20activation%20map. - @@ -1504,15 +1523,16 @@ - A layer that crops along the time dimension (axis 1) for 1D input. + A reshaping layer that crops along the time dimension (axis 1) for 1D input. + Cropping layer for 1D input (e.g. temporal sequence). It crops along the time dimension (axis 1). Cropping1D Layer - A layer that crops along the time dimension (axis 1) for 1D input. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Cropping1D + A reshaping layer that crops along the time dimension (axis 1) for 1D input. + @@ -1523,13 +1543,14 @@ A layer that crops along spatial dimensions (i.e. height and width) for 2D input. + Cropping layer for 2D input (e.g. picture). It crops along spatial dimensions, i.e. height and width. Cropping2D Layer A layer that crops along spatial dimensions (i.e. height and width) for 2D input. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Cropping2D + @@ -1540,13 +1561,14 @@ A layer that crops along spatial dimensions (depth + Cropping layer for 3D data (e.g. spatial or spatio-temporal). Cropping3D Layer A layer that crops along spatial dimensions (depth - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Cropping3D + @@ -1555,7 +1577,7 @@ - A LLM that performs well across a wide range of domains without significant loss in performance, facilitated by advanced domain adaptation techniques. + A large language model that performs well across a wide range of domains without significant loss in performance, facilitated by advanced domain adaptation techniques. Domain-General LLM cross-domain transfer domain adaptation @@ -1585,7 +1607,6 @@ A large language model that is trained by presenting learning examples in a meaningful order from simple to complex mimicking the learning trajectory followed by humans. - Curriculum Learning LLM Learning progression Curriculum Learning LLM @@ -1597,7 +1618,7 @@ - A technique used to increase the diversity and quantity of training data by applying various transformations such as rotation scaling flipping and cropping to existing data samples enhancing the robustness and performance of machine learning models. + A data enhancement used to increase the diversity and quantity of training data by applying various transformations such as rotation scaling flipping and cropping to existing data samples enhancing the robustness and performance of machine learning models. Data Enrichment Data Expansion Paraphrasing @@ -1621,8 +1642,7 @@ A use and interpretation bias where testing many hypotheses in a dataset may yield apparent statistical significance even when results are nonsignificant. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -1631,7 +1651,7 @@ - Techniques used to improve the quality diversity and volume of data available for training machine learning models such as data augmentation synthesis and enrichment to enhance model robustness and accuracy. + A preprocessing used to improve the quality diversity and volume of data available for training machine learning models, such as data augmentation synthesis and enrichment to enhance model robustness and accuracy. DataEnhancement @@ -1644,14 +1664,14 @@ A selection and sampling bias arising from adding synthetic or redundant data samples to a dataset. + Bias from adding synthetic or redundant data samples to a dataset. Data Generation Bias A selection and sampling bias arising from adding synthetic or redundant data samples to a dataset. - GTP-4o with Seppala et al. 2017 - https://en.wikipedia.org/wiki/Selection_bias + @@ -1659,16 +1679,17 @@ - + A machine learning task focused on methods that replace missing data with substituted values. + Methods that replace missing data with substituted values. Data Imputation A machine learning task focused on methods that replace missing data with substituted values. - https://en.wikipedia.org/wiki/Imputation_(statistics) + @@ -1677,7 +1698,7 @@ - The process of cleaning transforming and organizing raw data into a suitable format for analysis and modeling ensuring the quality and relevance of the data for machine learning tasks. + A preprocessing that cleans, transforms and organizes raw data into a suitable format for analysis and modeling, ensuring the quality and relevance of the data for machine learning tasks. Data Assembly Data Curation Data Processing @@ -1691,8 +1712,7 @@ - A LLM that generates natural language descriptions from structured data sources like tables, graphs, and knowledge bases, requiring grounding in meaning representations. - Data-to-Text LLM + A large language model that generates natural language descriptions from structured data sources like tables, graphs, and knowledge bases, requiring grounding in meaning representations. Meaning representation Data-to-Text LLM @@ -1704,15 +1724,16 @@ - A machine learning model that uses a tree-like model of decisions and their possible consequences including chance event outcomes resource costs and utilities. + A classification that uses a tree-like model of decisions and their possible consequences including chance event outcomes resource costs and utilities. + A decision support tool that uses a tree-like model of decisions and their possible consequences, including chance event outcomes, resource costs, and utilities. Decision Tree - A machine learning model that uses a tree-like model of decisions and their possible consequences including chance event outcomes resource costs and utilities. - https://en.wikipedia.org/wiki/Decision_tree + A classification that uses a tree-like model of decisions and their possible consequences including chance event outcomes resource costs and utilities. + @@ -1723,13 +1744,14 @@ A large language model that uses a decoder-only architecture consisting of only a decoder trained to predict the next token in a sequence given the previous tokens. + A decoder-only architecture consisting of only a decoder, trained to predict the next token in a sequence given the previous tokens. Unlike the encoder-decoder architecture, it does not have an explicit encoder and encodes information implicitly in the hidden state of the decoder, updated at each step of the generation process. Decoder LLM A large language model that uses a decoder-only architecture consisting of only a decoder trained to predict the next token in a sequence given the previous tokens. - https://www.practicalai.io/understanding-transformer-model-architectures/#:~:text=Encoder + @@ -1738,6 +1760,36 @@ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + A deep neural network that uses deconvolution for unsupervised construction of hierarchical image representations. DN @@ -1748,7 +1800,7 @@ A deep neural network that uses deconvolution for unsupervised construction of hierarchical image representations. - https://ieeexplore.ieee.org/document/5539957 + @@ -1766,7 +1818,7 @@ A deep neural network that combines deep learning and active learning to maximize model performance while annotating the fewest samples possible. - https://arxiv.org/pdf/2009.00236.pdf + @@ -1775,6 +1827,30 @@ + + + + + + + + + + + + + + + + + + + + + + + + An unsupervised pretrained network composed of multiple layers of latent variables that learns to probabilistically reconstruct inputs and perform classification. DBN @@ -1785,7 +1861,7 @@ An unsupervised pretrained network composed of multiple layers of latent variables that learns to probabilistically reconstruct inputs and perform classification. - https://en.wikipedia.org/wiki/Deep_belief_network + @@ -1794,6 +1870,42 @@ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + An autoencoder network that learns interpretable disentangled image representations through convolution and de-convolution layers trained with the stochastic gradient variational Bayes algorithm. DCIGN @@ -1807,6 +1919,42 @@ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + A deep neural network specialized for analyzing visual imagery using shared-weight architecture and translation-equivariant feature maps. CNN ConvNet @@ -1820,7 +1968,7 @@ A deep neural network specialized for analyzing visual imagery using shared-weight architecture and translation-equivariant feature maps. - https://en.wikipedia.org/wiki/Convolutional_neural_network + @@ -1829,6 +1977,24 @@ + + + + + + + + + + + + + + + + + + A deep neural network that processes information in one direction—from input nodes through hidden nodes to output nodes—without cycles or loops. DFF MLP @@ -1841,7 +2007,7 @@ A deep neural network that processes information in one direction—from input nodes through hidden nodes to output nodes—without cycles or loops. - https://en.wikipedia.org/wiki/Feedforward_neural_network + @@ -1853,6 +2019,7 @@ An artificial neural network characterized by multiple hidden layers between the input and output layers. DNN + A deep neural network (DNN) is a type of artificial neural network (ANN) characterized by multiple hidden layers between the input and output layers. Each layer consists of interconnected neurons that process and transmit information. DNNs can model complex patterns and representations in data through their hierarchical structure, where each layer extracts increasingly abstract features from the input. DNNs are widely used in various applications, including image and speech recognition, natural language processing, and more, due to their ability to learn and generalize from large amounts of data. Deep Neural Network @@ -1870,7 +2037,7 @@ A deep neural network that relaxes the hypothesis that training data must be independent and identically distributed with test data to address insufficient training data. - https://arxiv.org/abs/1808.01974 + @@ -1879,6 +2046,24 @@ + + + + + + + + + + + + + + + + + + An autoencoder network trained to reconstruct the original undistorted input from a partially corrupted input. DAE Denoising Autoencoder @@ -1890,7 +2075,7 @@ An autoencoder network trained to reconstruct the original undistorted input from a partially corrupted input. - https://doi.org/10.1145/1390156.1390294 + @@ -1901,13 +2086,14 @@ A layer that produces a dense tensor based on given feature columns. + A layer that produces a dense Tensor based on given feature_columns. Generally a single example in training data is described with FeatureColumns. At the first layer of the model, this column oriented data should be converted to a single Tensor. This layer can be called multiple times with different features. This is the V2 version of this layer that uses name_scopes to create variables instead of variable_scopes. But this approach currently lacks support for partitioned variables. In that case, use the V1 version instead. DenseFeatures Layer A layer that produces a dense tensor based on given feature columns. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/DenseFeatures + @@ -1918,13 +2104,14 @@ A layer that is a regular densely-connected neural network layer. + Just your regular densely-connected NN layer. Dense Layer A layer that is a regular densely-connected neural network layer. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Dense + @@ -1933,16 +2120,15 @@ - A bias arising when systems are used as decision aids for humans since the human intermediary may act on predictions in ways that are typically not modeled in the system. + A group bias arising when systems are used as decision aids for humans since the human intermediary may act on predictions in ways that are typically not modeled in the system. Deployment Bias - A bias arising when systems are used as decision aids for humans since the human intermediary may act on predictions in ways that are typically not modeled in the system. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + A group bias arising when systems are used as decision aids for humans since the human intermediary may act on predictions in ways that are typically not modeled in the system. + @@ -1951,15 +2137,16 @@ - A layer that performs depthwise 1D convolution + A convolutional layer that performs depthwise 1D convolution + Depthwise 1D convolution. Depthwise convolution is a type of convolution in which each input channel is convolved with a different kernel (called a depthwise kernel). You can understand depthwise convolution as the first step in a depthwise separable convolution. It is implemented via the following steps: Split the input into individual channels. Convolve each channel with an individual depthwise kernel with depth_multiplier output channels. Concatenate the convolved outputs along the channels axis. Unlike a regular 1D convolution, depthwise convolution does not mix information across different input channels. The depth_multiplier argument determines how many filter are applied to one input channel. As such, it controls the amount of output channels that are generated per input channel in the depthwise step. DepthwiseConv1D Layer - A layer that performs depthwise 1D convolution - https://www.tensorflow.org/api_docs/python/tf/keras/layers/DepthwiseConv1D + A convolutional layer that performs depthwise 1D convolution + @@ -1968,15 +2155,16 @@ - A layer that performs depthwise 2D convolution + A convolutional layer that performs depthwise 2D convolution + Depthwise 2D convolution. DepthwiseConv2D Layer - A layer that performs depthwise 2D convolution - https://www.tensorflow.org/api_docs/python/tf/keras/layers/DepthwiseConv2D + A convolutional layer that performs depthwise 2D convolution + @@ -1987,14 +2175,14 @@ A selection and sampling bias characterized by systematic differences between groups in how outcomes are determined potentially over- or underestimating effect size. + Systematic differences between groups in how outcomes are determined, potentially over- or underestimating effect size. Detection Bias A selection and sampling bias characterized by systematic differences between groups in how outcomes are determined potentially over- or underestimating effect size. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -2031,7 +2219,7 @@ - A machine learning task focused on the process of transforming data from a high-dimensional space into a lower-dimensional space while retaining meaningful properties of the original data. + An unsupervised learning focused on the process of transforming data from a high-dimensional space into a lower-dimensional space while retaining meaningful properties of the original data. Dimension Reduction Dimensionality Reduction @@ -2039,8 +2227,8 @@ - A machine learning task focused on the process of transforming data from a high-dimensional space into a lower-dimensional space while retaining meaningful properties of the original data. - https://en.wikipedia.org/wiki/Dimensionality_reduction + An unsupervised learning focused on the process of transforming data from a high-dimensional space into a lower-dimensional space while retaining meaningful properties of the original data. + @@ -2049,15 +2237,15 @@ - A preprocessing layer which buckets continuous features by ranges. + A numerical features prepreprocessing layer which buckets continuous features by ranges. Discretization Layer - A preprocessing layer which buckets continuous features by ranges. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Discretization + A numerical features prepreprocessing layer which buckets continuous features by ranges. + @@ -2066,7 +2254,7 @@ - The process of training a smaller model to replicate the behavior of a larger model aiming to compress the knowledge into a more compact form without significant loss of performance. + A preprocessing that trains a smaller model to replicate the behavior of a larger model aiming to compress the knowledge into a more compact form without significant loss of performance. Purification Refining Knowledge compression @@ -2077,8 +2265,8 @@ - The process of training a smaller model to replicate the behavior of a larger model aiming to compress the knowledge into a more compact form without significant loss of performance. - https://doi.org/10.48550/arXiv.2105.13093 + A preprocessing that trains a smaller model to replicate the behavior of a larger model aiming to compress the knowledge into a more compact form without significant loss of performance. + @@ -2087,7 +2275,7 @@ - A LLM which is pre-trained on a broad corpus and then fine-tuned on domain-specific data to specialize its capabilities for particular domains or applications, like scientific literature or code generation. + A large language model which is pre-trained on a broad corpus and then fine-tuned on domain-specific data to specialize its capabilities for particular domains or applications, like scientific literature or code generation. Domain-Adapted Large Language Model domain robustness transfer learning @@ -2103,13 +2291,14 @@ A layer that computes a dot product between samples in two tensors. + Layer that computes a dot product between samples in two tensors. E.g. if applied to a list of two tensors a and b of shape (batch_size, n), the output will be a tensor of shape (batch_size, 1) where each entry i will be the dot product between a[i] and b[i]. Dot Layer A layer that computes a dot product between samples in two tensors. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Dot + @@ -2120,13 +2309,14 @@ A regularization layer that applies Dropout to the input + Applies Dropout to the input. The Dropout layer randomly sets input units to 0 with a frequency of rate at each step during training time, which helps prevent overfitting. Inputs not set to 0 are scaled up by 1/(1 - rate) such that the sum over all inputs is unchanged. Note that the Dropout layer only applies when training is set to True such that no values are dropped during inference. When using model.fit, training will be appropriately set to True automatically, and in other contexts, you can set the kwarg explicitly to True when calling the layer. (This is in contrast to setting trainable=False for a Dropout layer. trainable does not affect the layer's behavior, as Dropout does not have any variables/weights that can be frozen during training.) Dropout Layer A regularization layer that applies Dropout to the input - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Dropout + @@ -2144,8 +2334,7 @@ A cognitive bias in which people with low ability in an area overestimate that ability. Often measured by comparing self-assessment with objective performance. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -2165,18 +2354,20 @@ - - An activation function that is x if x > 0 and alpha * (exp(x) - 1) if x < 0 where alpha controls the value to which an ELU saturates for negative net inputs. + + A mathematical function that is x if x > 0 and alpha * (exp(x) - 1) if x < 0 where alpha controls the value to which an ELU saturates for negative net inputs. ELU Exponential Linear Unit + + The exponential linear unit (ELU) with alpha > 0 is: x if x > 0 and alpha * (exp(x) - 1) if x < 0 The ELU hyperparameter alpha controls the value to which an ELU saturates for negative net inputs. ELUs diminish the vanishing gradient effect. ELUs have negative values which pushes the mean of the activations closer to zero. Mean activations that are closer to zero enable faster Learning as they bring the gradient closer to the natural gradient. ELUs saturate to a negative value when the argument gets smaller. Saturation means a small derivative which decreases the variation and the information that is propagated to the next layer. ELU Function - An activation function that is x if x > 0 and alpha * (exp(x) - 1) if x < 0 where alpha controls the value to which an ELU saturates for negative net inputs. - https://www.tensorflow.org/api_docs/python/tf/keras/activations/elu + A mathematical function that is x if x > 0 and alpha * (exp(x) - 1) if x < 0 where alpha controls the value to which an ELU saturates for negative net inputs. + @@ -2187,13 +2378,14 @@ An activation layer that applies the Exponential Linear Unit (ELU) function element-wise. + Exponential Linear Unit. ELU Layer An activation layer that applies the Exponential Linear Unit (ELU) function element-wise. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/ELU + @@ -2202,6 +2394,24 @@ + + + + + + + + + + + + + + + + + + A recurrent neural network with a recurrent hidden layer and sparsely connected hidden neurons that learns output weights to produce temporal patterns. ESN @@ -2212,7 +2422,7 @@ A recurrent neural network with a recurrent hidden layer and sparsely connected hidden neurons that learns output weights to produce temporal patterns. - https://en.wikipedia.org/wiki/Echo_state_network#:~:text=The%20echo%20state%20network%20(ESN,are%20fixed%20and%20randomly%20assigned + @@ -2230,8 +2440,7 @@ A selection and sampling bias occurring when an inference about an individual is made based on their group membership. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -2248,7 +2457,7 @@ A layer that turns positive integers (indexes) into dense vectors of fixed size. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Embedding + @@ -2261,6 +2470,7 @@ Embodied Large Language Model multimodal grounding + An embodied LLM integrates language with other modalities like vision, audio, and robotics to enable grounded language understanding in real-world environments. Embodied LLM @@ -2278,8 +2488,7 @@ A use and interpretation bias resulting from the use and reliance on algorithms across new or unanticipated contexts. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -2288,15 +2497,15 @@ - The LLM introduced in the "Attention Is All You Need" paper. The encoder processes the input sequence to generate a hidden representation summarizing the input information, while the decoder uses this hidden representation to generate the desired output sequence. + A large language model introduced in the "Attention Is All You Need" paper. The encoder processes the input sequence to generate a hidden representation summarizing the input information, while the decoder uses this hidden representation to generate the desired output sequence. Encoder-Decoder LLM - The LLM introduced in the "Attention Is All You Need" paper. The encoder processes the input sequence to generate a hidden representation summarizing the input information, while the decoder uses this hidden representation to generate the desired output sequence. - https://www.practicalai.io/understanding-transformer-model-architectures/#:~:text=Encoder + A large language model introduced in the "Attention Is All You Need" paper. The encoder processes the input sequence to generate a hidden representation summarizing the input information, while the decoder uses this hidden representation to generate the desired output sequence. + @@ -2307,13 +2516,14 @@ A large language model that uses an encoder-only architecture to encode the input sequence into a fixed-length representation which is then used as input to a classifier or regressor for prediction. + An encoder-only architecture that encodes the input sequence into a fixed-length representation, which is then used as input to a classifier or regressor for prediction. The model has a pre-trained general-purpose encoder that requires fine-tuning for specific tasks. Encoder LLM A large language model that uses an encoder-only architecture to encode the input sequence into a fixed-length representation which is then used as input to a classifier or regressor for prediction. - https://www.practicalai.io/understanding-transformer-model-architectures/#:~:text=Encoder + @@ -2322,7 +2532,7 @@ - A LLM which models the explicit probability density over token sequences using an energy function, rather than an autoregressive factorization. This can improve modeling of long-range dependencies and global coherence. + A large language model which models the explicit probability density over token sequences using an energy function, rather than an autoregressive factorization. This can improve modeling of long-range dependencies and global coherence. Energy-Based Large Language Model energy scoring explicit density modeling @@ -2335,16 +2545,16 @@ - - A type of machine learning focused on methods that use multiple learning algorithms to achieve better predictive performance than any of the constituent algorithms alone. + + A machine learning task focused on methods that use multiple learning algorithms to achieve better predictive performance than any of the constituent algorithms alone. Ensemble Learning - A type of machine learning focused on methods that use multiple learning algorithms to achieve better predictive performance than any of the constituent algorithms alone. - https://en.wikipedia.org/wiki/Ensemble_learning + A machine learning task focused on methods that use multiple learning algorithms to achieve better predictive performance than any of the constituent algorithms alone. + @@ -2362,8 +2572,7 @@ A processing bias characterized by the effect of variables' uncertainties (or errors more specifically random errors) on the uncertainty of a function based on them. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -2394,8 +2603,7 @@ A selection and sampling bias arising when testing populations do not equally represent user populations or when inappropriate performance metrics are used. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -2426,8 +2634,7 @@ A selection and sampling bias occurring when specific groups of user populations are excluded from testing and analysis. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -2450,16 +2657,17 @@ - - An activation function that is the mathematical function denoted by f(x)=exp or e^{x}. + + A mathematical function denoted by f(x)=exp or e^{x}. + The exponential function is a mathematical function denoted by f(x)=exp or e^{x}. Exponential Function - An activation function that is the mathematical function denoted by f(x)=exp or e^{x}. - https://www.tensorflow.org/api_docs/python/tf/keras/activations/exponential + A mathematical function denoted by f(x)=exp or e^{x}. + @@ -2480,6 +2688,24 @@ + + + + + + + + + + + + + + + + + + A feedback network with randomly assigned hidden nodes that are not updated during training. ELM @@ -2490,7 +2716,7 @@ A feedback network with randomly assigned hidden nodes that are not updated during training. - https://en.wikipedia.org/wiki/Extreme_Learning_machine + @@ -2508,7 +2734,7 @@ A language model that views each word as a vector of multiple factors such as part-of-speech morphology and semantics to improve language modeling. - https://en.wikipedia.org/wiki/Factored_language_model + @@ -2529,7 +2755,7 @@ A large language model that decomposes the full language modeling task into multiple sub-components or experts that each focus on a subset of the information enabling more efficient scaling. - https://doi.org/10.48550/arXiv.2403.12556 + @@ -2538,7 +2764,7 @@ - The process of transforming raw data into a set of measurable characteristics that can be used as input for machine learning algorithms enhancing the ability to make accurate predictions. + A data enhancement that transforms raw data into a set of measurable characteristics that can be used as input for machine learning algorithms, enhancing the ability to make accurate predictions. Attribute Extraction Feature Isolation Semantic embeddings @@ -2569,13 +2795,14 @@ A deep neural network trained across decentralized edge devices or servers holding local data samples without exchanging them. + Training an algorithm across multiple decentralized edge devices or servers holding local data samples without exchanging them. Federated Learning A deep neural network trained across decentralized edge devices or servers holding local data samples without exchanging them. - https://en.wikipedia.org/wiki/Federated_learning + @@ -2592,8 +2819,7 @@ A use and interpretation bias occurring when an algorithm learns from user behavior and feeds that behavior back into the model. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -2602,6 +2828,24 @@ + + + + + + + + + + + + + + + + + + An artificial neural network that refines its representations iteratively based on feedback from previous outputs. FBN @@ -2615,7 +2859,7 @@ - A regression analysis model in which the model parameters are fixed or non-random quantities. + A regression analysis in which the model parameters are fixed or non-random quantities. FEM Fixed Effects Model @@ -2623,8 +2867,8 @@ - A regression analysis model in which the model parameters are fixed or non-random quantities. - https://en.wikipedia.org/wiki/Fixed_effects_model + A regression analysis in which the model parameters are fixed or non-random quantities. + @@ -2633,15 +2877,16 @@ - A layer that flattens the input + A reshaping layer that flattens the input + Flattens the input. Does not affect the batch size. Flatten Layer - A layer that flattens the input - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Flatten + A reshaping layer that flattens the input + @@ -2652,7 +2897,6 @@ A pooling layer that applies a 2D fractional max pooling over an input signal composed of several input planes. FractionalMaxPool2D - FractionalMaxPool2d FractionalMaxPool2D Layer @@ -2660,7 +2904,7 @@ A pooling layer that applies a 2D fractional max pooling over an input signal composed of several input planes. - https://pytorch.org/docs/stable/nn.html#pooling-layers + @@ -2671,7 +2915,6 @@ A pooling layer that applies a 3D fractional max pooling over an input signal composed of several input planes. FractionalMaxPool3D - FractionalMaxPool3d FractionalMaxPool3D Layer @@ -2679,24 +2922,7 @@ A pooling layer that applies a 3D fractional max pooling over an input signal composed of several input planes. - https://pytorch.org/docs/stable/nn.html#pooling-layers - - - - - - - - - A mathematical rule that gives the value of a dependent variable corresponding to specified values of independent variables. - - Function - - - - - A mathematical rule that gives the value of a dependent variable corresponding to specified values of independent variables. - https://www.sciencedirect.com/topics/mathematics/mathematical-function + @@ -2705,16 +2931,15 @@ - A bias arising when biased results are reported to support or satisfy the funding agency or financial supporter of a research study. + A group bias arising when biased results are reported to support or satisfy the funding agency or financial supporter of a research study. Funding Bias - A bias arising when biased results are reported to support or satisfy the funding agency or financial supporter of a research study. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + A group bias arising when biased results are reported to support or satisfy the funding agency or financial supporter of a research study. + @@ -2722,18 +2947,19 @@ - - An activation function that computes x * P(X <= x) where P(X) ~ N(0 1) weighting inputs by their value rather than gating inputs by their sign as in ReLU. + + A mathematical function that computes x * P(X <= x) where P(X) ~ N(0 1) weighting inputs by their value rather than gating inputs by their sign as in ReLU. GELU Gaussian Error Linear Unit + Gaussian error linear unit (GELU) computes x * P(X <= x), where P(X) ~ N(0, 1). The (GELU) nonlinearity weights inputs by their value, rather than gates inputs by their sign as in ReLU. GELU Function - An activation function that computes x * P(X <= x) where P(X) ~ N(0 1) weighting inputs by their value rather than gating inputs by their sign as in ReLU. - https://www.tensorflow.org/api_docs/python/tf/keras/activations/gelu + A mathematical function that computes x * P(X <= x) where P(X) ~ N(0 1) weighting inputs by their value rather than gating inputs by their sign as in ReLU. + @@ -2744,13 +2970,14 @@ A layer that processes one step within the whole time sequence input for a GRU layer. + Cell class for the GRU layer. This class processes one step within the whole time sequence input, whereas tf.keras.layer.GRU processes the whole sequence. GRUCell Layer A layer that processes one step within the whole time sequence input for a GRU layer. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/GRUCell + @@ -2761,13 +2988,14 @@ A recurrent layer that implements the Gated Recurrent Unit architecture. + Gated Recurrent Unit - Cho et al. 2014. Based on available runtime hardware and constraints, this layer will choose different implementations (cuDNN-based or pure-TensorFlow) to maximize the performance. If a GPU is available and all the arguments to the layer meet the requirement of the cuDNN kernel (see below for details), the layer will use a fast cuDNN implementation. The requirements to use the cuDNN implementation are: activation == tanh, recurrent_activation == sigmoid, recurrent_dropout == 0, unroll is False, use_bias is True, reset_after is True. Inputs, if use masking, are strictly right-padded. Eager execution is enabled in the outermost context. There are two variants of the GRU implementation. The default one is based on v3 and has reset gate applied to hidden state before matrix multiplication. The other one is based on original and has the order reversed. The second variant is compatible with CuDNNGRU (GPU-only) and allows inference on CPU. Thus it has separate biases for kernel and recurrent_kernel. To use this variant, set reset_after=True and recurrent_activation='sigmoid'. GRU Layer A recurrent layer that implements the Gated Recurrent Unit architecture. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/GRU + @@ -2776,6 +3004,24 @@ + + + + + + + + + + + + + + + + + + A long short-term memory network that is a gating mechanism in recurrent neural networks similar to LSTMs but with fewer parameters and no output gate. GRU @@ -2786,7 +3032,7 @@ A long short-term memory network that is a gating mechanism in recurrent neural networks similar to LSTMs but with fewer parameters and no output gate. - https://en.wikipedia.org/wiki/Gated_recurrent_unit + @@ -2797,13 +3043,14 @@ A regularization layer that applies multiplicative 1-centered Gaussian noise. + Apply multiplicative 1-centered Gaussian noise. As it is a regularization layer, it is only active at training time. GaussianDropout Layer A regularization layer that applies multiplicative 1-centered Gaussian noise. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/GaussianDropout + @@ -2814,13 +3061,14 @@ A regularization layer that applies additive zero-centered Gaussian noise. + Apply additive zero-centered Gaussian noise. This is useful to mitigate overfitting (you could see it as a form of random data augmentation). Gaussian Noise (GS) is a natural choice as corruption process for real valued inputs. As it is a regularization layer, it is only active at training time. GaussianNoise Layer A regularization layer that applies additive zero-centered Gaussian noise. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/GaussianNoise + @@ -2850,7 +3098,7 @@ A deep neural network that learns novel classes from few samples per class, preventing catastrophic forgetting of base classes and ensuring classifier calibration. - https://paperswithcode.com/paper/generalized-and-incremental-few-shot-learning/review/ + @@ -2859,7 +3107,7 @@ - A machine learning model that generalizes linear regression by relating the linear model to the response variable via a link function and allowing the variance of each measurement to be a function of its predicted value. + A regression analysis that relates the linear model to the response variable via a link function and allowing the variance of each measurement to be a function of its predicted value. GLM Generalized Linear Model @@ -2867,8 +3115,8 @@ - A machine learning model that generalizes linear regression by relating the linear model to the response variable via a link function and allowing the variance of each measurement to be a function of its predicted value. - https://en.wikipedia.org/wiki/Generalized_linear_model + A regression analysis that relates the linear model to the response variable via a link function and allowing the variance of each measurement to be a function of its predicted value. + @@ -2877,6 +3125,24 @@ + + + + + + + + + + + + + + + + + + An unsupervised pretrained network framework where two neural networks contest in a game to generate new data with the same statistics as the training set. GAN @@ -2887,7 +3153,7 @@ An unsupervised pretrained network framework where two neural networks contest in a game to generate new data with the same statistics as the training set. - https://en.wikipedia.org/wiki/Generative_adversarial_network + @@ -2896,7 +3162,7 @@ - A LLM which incorporates a generative adversarial network (GAN) into its training process, using a discriminator network to provide a signal for generating more realistic and coherent text. This adversarial training can improve the quality and diversity of generated text. + A large language model which incorporates a generative adversarial network (GAN) into its training process, using a discriminator network to provide a signal for generating more realistic and coherent text. This adversarial training can improve the quality and diversity of generated text. GAN-Large Language Model Generative Adversarial Network-Augmented Large Language Model adversarial training @@ -2923,7 +3189,7 @@ A large language model that is trained to understand and model basic physics causality and common sense about how the real world works. - https://arxiv.org/abs/2306.12672 + @@ -2933,7 +3199,6 @@ A language model that enables users to engage in an interactive dialogue with an LLM providing feedback to guide and refine the generated outputs iteratively. - Generative Language Interface Interactive generation Generative Language Interface @@ -2948,13 +3213,14 @@ A pooling layer that performs global average pooling operation for temporal data. GlobalAvgPool1D + Global average pooling operation for temporal data. GlobalAveragePooling1D Layer A pooling layer that performs global average pooling operation for temporal data. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalAveragePooling1D + @@ -2966,13 +3232,14 @@ A pooling layer that performs global average pooling operation for spatial data. GlobalAvgPool2D + Global average pooling operation for spatial data. GlobalAveragePooling2D Layer A pooling layer that performs global average pooling operation for spatial data. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalAveragePooling2D + @@ -2984,13 +3251,14 @@ A pooling layer that performs global average pooling operation for 3D data. GlobalAvgPool3D + Global Average pooling operation for 3D data. GlobalAveragePooling3D Layer A pooling layer that performs global average pooling operation for 3D data. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalAveragePooling3D + @@ -3002,13 +3270,14 @@ A pooling layer that performs global max pooling operation for temporal data. GlobalMaxPool1D + Global max pooling operation for 1D temporal data. GlobalMaxPooling1D Layer A pooling layer that performs global max pooling operation for temporal data. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalMaxPool1D + @@ -3020,13 +3289,14 @@ A pooling layer that performs global max pooling operation for spatial data. GlobalMaxPool2D + Global max pooling operation for spatial data. GlobalMaxPooling2D Layer A pooling layer that performs global max pooling operation for spatial data. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalMaxPool2D + @@ -3038,13 +3308,14 @@ A pooling layer that performs global max pooling operation for 3D data. GlobalMaxPool3D + Global Max pooling operation for 3D data. GlobalMaxPooling3D Layer A pooling layer that performs global max pooling operation for 3D data. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/GlobalMaxPool3D + @@ -3053,6 +3324,24 @@ + + + + + + + + + + + + + + + + + + A deep neural network that operates directly on graph structures utilizing structural information. GCN @@ -3063,7 +3352,7 @@ A deep neural network that operates directly on graph structures utilizing structural information. - https://arxiv.org/abs/1609.02907 + @@ -3072,6 +3361,30 @@ + + + + + + + + + + + + + + + + + + + + + + + + A graph convolutional network that generates goal-directed graphs using reinforcement learning and optimizing for rewards and adversarial loss. GPCN @@ -3082,7 +3395,7 @@ A graph convolutional network that generates goal-directed graphs using reinforcement learning and optimizing for rewards and adversarial loss. - https://arxiv.org/abs/1806.02473 + @@ -3101,7 +3414,7 @@ A language model that operates over structured inputs or outputs represented as graphs enabling reasoning over explicit relational knowledge representations during language tasks. - https://arxiv.org/abs/2401.07105 + @@ -3109,22 +3422,22 @@ - - A bias characterized by favoring members of one's in-group over out-group members expressed in evaluation resource allocation and other ways. + + A systemic bias characterized by favoring members of one's in-group over out-group members expressed in evaluation resource allocation and other ways. In-group Favoritism In-group bias In-group preference In-group–out-group Bias Intergroup bias + Favoring members of one's in-group over out-group members, expressed in evaluation, resource allocation, and other ways. Group Bias - A bias characterized by favoring members of one's in-group over out-group members expressed in evaluation resource allocation and other ways. - GTP-4o with Seppala et al. 2017 - https://en.wikipedia.org/wiki/In-group_favoritism + A systemic bias characterized by favoring members of one's in-group over out-group members expressed in evaluation resource allocation and other ways. + @@ -3136,13 +3449,14 @@ A normalization layer that applies Group Normalization over a mini-batch of inputs. GroupNorm + Applies Group Normalization over a mini-batch of inputs as described in the paper Group Normalization GroupNorm Layer A normalization layer that applies Group Normalization over a mini-batch of inputs. - https://pytorch.org/docs/stable/nn.html#normalization-layers + @@ -3151,7 +3465,7 @@ - A psychological phenomenon where people in a group make non-optimal decisions due to a desire to conform or fear of dissent. + A group bias in which people in a group make non-optimal decisions due to a desire to conform or fear of dissent. Groupthink Groupthink Bias @@ -3159,9 +3473,8 @@ - A psychological phenomenon where people in a group make non-optimal decisions due to a desire to conform or fear of dissent. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + A group bias in which people in a group make non-optimal decisions due to a desire to conform or fear of dissent. + @@ -3169,16 +3482,17 @@ - - An activation function that is a faster approximation of the sigmoid activation using a piecewise linear approximation. + + A mathematical function that is a faster approximation of the sigmoid activation using a piecewise linear approximation. + A faster approximation of the sigmoid activation. Piecewise linear approximation of the sigmoid function. Ref: 'https://en.wikipedia.org/wiki/Hard_sigmoid' Hard Sigmoid Function - An activation function that is a faster approximation of the sigmoid activation using a piecewise linear approximation. - https://www.tensorflow.org/api_docs/python/tf/keras/activations/hard_sigmoid + A mathematical function that is a faster approximation of the sigmoid activation using a piecewise linear approximation. + @@ -3189,13 +3503,14 @@ A categorical features preprocessing layer which hashes and bins categorical features. + A preprocessing layer which hashes and bins categorical features. This layer transforms categorical inputs to hashed output. It element-wise converts a ints or strings to ints in a fixed range. The stable hash function uses tensorflow::ops::Fingerprint to produce the same output consistently across all platforms. This layer uses FarmHash64 by default, which provides a consistent hashed output across different platforms and is stable across invocations, regardless of device and context, by mixing the input bits thoroughly. If you want to obfuscate the hashed output, you can also pass a random salt argument in the constructor. In that case, the layer will use the SipHash64 hash function, with the salt value serving as additional input to the hash function. Hashing Layer A categorical features preprocessing layer which hashes and bins categorical features. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Hashing + @@ -3204,16 +3519,7 @@ - A layer located between the input and output that performs nonlinear transformations of the inputs entered into the network. - - Hidden Layer - - - - A layer located between the input and output that performs nonlinear transformations of the inputs entered into the network. - https://deepai.org/machine-Learning-glossary-and-terms/hidden-layer-machine-Learning - @@ -3221,15 +3527,16 @@ - A classification task focused on methods that group things according to a hierarchy. + A classification focused on methods that group things according to a hierarchy. + Methods that group things according to a hierarchy. Hierarchical Classification - A classification task focused on methods that group things according to a hierarchy. - https://en.wikipedia.org/wiki/Hierarchical_classification + A classification focused on methods that group things according to a hierarchy. + @@ -3238,16 +3545,17 @@ - A clustering method that builds a hierarchy of clusters. + A clustering that builds a hierarchy of clusters. HCL + Methods that build a hierarchy of clusters. Hierarchical Clustering - A clustering method that builds a hierarchy of clusters. - https://en.wikipedia.org/wiki/Hierarchical_clustering + A clustering that builds a hierarchy of clusters. + @@ -3266,7 +3574,7 @@ A language model that represents language at multiple levels of granularity learning hierarchical representations that capture both low-level patterns and high-level abstractions. - https://doi.org/10.1016/j.ipm.2024.103698 + @@ -3277,14 +3585,14 @@ A bias characterized by long-standing biases encoded in society over time distinct from biases in historical description or interpretation. + Long-standing biases encoded in society over time, distinct from biases in historical description or the interpretation of history, such as viewing the larger world from a Western or European perspective. Historical Bias A bias characterized by long-standing biases encoded in society over time distinct from biases in historical description or interpretation. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -3293,6 +3601,12 @@ + + + + + + A symmetrically connected network that is a type of recurrent artificial neural network serving as a content-addressable memory system. HN Ising model of a neural network @@ -3305,7 +3619,7 @@ A symmetrically connected network that is a type of recurrent artificial neural network serving as a content-addressable memory system. - https://en.wikipedia.org/wiki/Hopfield_network + @@ -3316,14 +3630,14 @@ A use and interpretation bias where individuals perceive benign or ambiguous behaviors as hostile. + Bias where individuals perceive benign or ambiguous behaviors as hostile. Hostile Attribution Bias A use and interpretation bias where individuals perceive benign or ambiguous behaviors as hostile. - GTP-4o with Seppala et al. 2017 - https://en.wikipedia.org/wiki/Interpretive_bias + @@ -3332,16 +3646,15 @@ - A systematic error in human thought based on heuristic principles leading to simplified judgmental operations. + A bias in human thought based on heuristic principles leading to simplified judgmental operations. Human Bias - A systematic error in human thought based on heuristic principles leading to simplified judgmental operations. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + A bias in human thought based on heuristic principles leading to simplified judgmental operations. + @@ -3358,8 +3671,7 @@ An individual bias that arises when users depend on automated systems as heuristic substitutes for their own information-seeking and processing efforts. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -3376,7 +3688,7 @@ A layer that performs image data preprocessing augmentations. - https://keras.io/guides/preprocessing_layers/ + @@ -3393,7 +3705,7 @@ A layer that performs image data preprocessing operations. - https://keras.io/guides/preprocessing_layers/ + @@ -3405,14 +3717,14 @@ An individual bias characterized by unconscious beliefs attitudes feelings associations or stereotypes that affect information processing decision-making and actions. Confirmatory Bias + Unconscious beliefs, attitudes, feelings, associations, or stereotypes that affect information processing, decision-making, and actions. Implicit Bias An individual bias characterized by unconscious beliefs attitudes feelings associations or stereotypes that affect information processing decision-making and actions. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -3432,7 +3744,7 @@ A language model that uses an energy function to score entire sequences instead of factorizing probabilities autoregressively better capturing global properties and long-range dependencies. - https://arxiv.org/pdf/2303.16189 + @@ -3450,7 +3762,7 @@ A deep neural network trained on a base set of classes and then presented with novel classes, each with few labeled examples. - https://arxiv.org/abs/1810.07218 + @@ -3459,16 +3771,15 @@ - A persistent point of view or limited list of such points of view applied by an individual. + A bias characterized by a persistent point of view or limited list of such points of view, applied by an individual. Individual Bias - A persistent point of view or limited list of such points of view applied by an individual. - GTP-4o with Seppala et al. 2017 - https://develop.consumerium.org/wiki/Individual_bias + A bias characterized by a persistent point of view or limited list of such points of view, applied by an individual. + @@ -3485,8 +3796,7 @@ A processing bias arising when machine learning applications generate inputs for other machine learning algorithms passing on any existing bias. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -3495,16 +3805,7 @@ - A layer composed of artificial input neurons that brings the initial data into the system for further processing by subsequent layers. - - Input Layer - - - - A layer composed of artificial input neurons that brings the initial data into the system for further processing by subsequent layers. - https://www.techopedia.com/definition/33262/input-layer-neural-networks#:~:text=Explains%20Input%20Layer-,What%20Does%20Input%20Layer%20Mean%3F,for%20the%20artificial%20neural%20network. - @@ -3520,7 +3821,7 @@ A layer to be used as an entry point into a Network (a graph of layers). - https://www.tensorflow.org/api_docs/python/tf/keras/layers/InputLayer + @@ -3531,70 +3832,70 @@ A layer that specifies the rank + Specifies the rank, dtype and shape of every input to a layer. Layers can expose (if appropriate) an input_spec attribute: an instance of InputSpec, or a nested structure of InputSpec instances (one per input tensor). These objects enable the layer to run input compatibility checks for input structure, input rank, input shape, and input dtype. A None entry in a shape is compatible with any dimension, a None shape is compatible with any shape. InputSpec Layer A layer that specifies the rank - https://www.tensorflow.org/api_docs/python/tf/keras/layers/InputSpec + - + - + A normalization layer that applies Instance Normalization over a 2D (unbatched) or 3D (batched) input. InstanceNorm1D - InstanceNorm1d - InstanceNorm1d Layer + Applies Instance Normalization over a 2D (unbatched) or 3D (batched) input as described in the paper Instance Normalization: The Missing Ingredient for Fast Stylization. + InstanceNorm1D Layer - + A normalization layer that applies Instance Normalization over a 2D (unbatched) or 3D (batched) input. - https://pytorch.org/docs/stable/nn.html#normalization-layers + - + - + A normalization layer that applies Instance Normalization over a 4D input (a mini-batch of 2D inputs with additional channel dimension). - InstanceNorm2D - InstanceNorm2d - InstanceNorm2d + Applies Instance Normalization over a 4D input (a mini-batch of 2D inputs with additional channel dimension) as described in the paper Instance Normalization: The Missing Ingredient for Fast Stylization. + InstanceNorm2D - + A normalization layer that applies Instance Normalization over a 4D input (a mini-batch of 2D inputs with additional channel dimension). - https://pytorch.org/docs/stable/nn.html#normalization-layers + - + - + A normalization layer that applies Instance Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension). InstanceNorm3D - InstanceNorm3d - InstanceNorm3d Layer + Applies Instance Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension) as described in the paper Instance Normalization: The Missing Ingredient for Fast Stylization. + InstanceNorm3D Layer - + A normalization layer that applies Instance Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension). - https://pytorch.org/docs/stable/nn.html#normalization-layers + @@ -3602,17 +3903,17 @@ - - A bias exhibited at the level of entire institutions where practices or norms result in the favoring or disadvantaging of certain social groups. + + A systemic bias exhibited at the level of entire institutions where practices or norms result in the favoring or disadvantaging of certain social groups. + Bias exhibited at the level of entire institutions, where practices or norms result in the favoring or disadvantaging of certain social groups, such as institutional racism or sexism. Institutional Bias - A bias exhibited at the level of entire institutions where practices or norms result in the favoring or disadvantaging of certain social groups. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + A systemic bias exhibited at the level of entire institutions where practices or norms result in the favoring or disadvantaging of certain social groups. + @@ -3621,7 +3922,7 @@ - A LLM which is fine-tuned to follow natural language instructions accurately and safely, learning to map from instructions to desired model behavior in a more controlled and principled way. + A large language model which is fine-tuned to follow natural language instructions accurately and safely, learning to map from instructions to desired model behavior in a more controlled and principled way. Instruction-Tuned Large Language Model constitutional AI natural language instructions @@ -3643,7 +3944,7 @@ A categorical features preprocessing layer that maps integer features to contiguous ranges. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/IntegerLookup + @@ -3660,8 +3961,7 @@ An individual bias where users interpret algorithmic outputs according to their internalized biases and views. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -3670,9 +3970,6 @@ - A layer that obtains the dot product of input values or subsets of input values. - - Kernel Layer @@ -3680,8 +3977,8 @@ - - A machine learning that groups objects by a plurality vote of its neighbors, assigning each object to the class most common among its k nearest neighbors. + + A machine learning task that groups objects by a plurality vote of its neighbors, assigning each object to the class most common among its k nearest neighbors. K-NN KNN @@ -3690,8 +3987,8 @@ - A machine learning that groups objects by a plurality vote of its neighbors, assigning each object to the class most common among its k nearest neighbors. - https://en.wikipedia.org/wiki/K-nearest_neighbors_algorithm + A machine learning task that groups objects by a plurality vote of its neighbors, assigning each object to the class most common among its k nearest neighbors. + @@ -3699,11 +3996,10 @@ - A classification and clustering that classifies objects by a plurality vote of its neighbors, assigning each object to the class most common among its k nearest neighbors. - K-NN - KNN + K-NN Classification + KNN Classification K-nearest Neighbor Classification Algorithm @@ -3711,7 +4007,7 @@ A classification and clustering that classifies objects by a plurality vote of its neighbors, assigning each object to the class most common among its k nearest neighbors. - https://en.wikipedia.org/wiki/K-nearest_neighbors_algorithm + @@ -3721,8 +4017,8 @@ An regression analysis that assigns the average of the values of k nearest neighbors to objects. - K-NN - KNN + K-NN Regression + KNN Regression K-nearest Neighbor Regression Algorithm @@ -3730,7 +4026,7 @@ An regression analysis that assigns the average of the values of k nearest neighbors to objects. - https://en.wikipedia.org/wiki/K-nearest_neighbors_algorithm + @@ -3739,7 +4035,7 @@ - A LLM which incorporates external knowledge sources or knowledge bases into the model architecture, enabling it to generate more factually accurate and knowledge-aware text. + A large language model which incorporates external knowledge sources or knowledge bases into the model architecture, enabling it to generate more factually accurate and knowledge-aware text. Knowledge-Grounded Large Language Model factual grounding knowledge integration @@ -3753,7 +4049,7 @@ - The process by which knowledge is passed from one entity such as a person organization or system to another facilitating learning and adaptation in the receiving entity through various methods such as teaching training or data exchange. + A training strategy in which knowledge is passed from one entity such as a person organization or system to another facilitating learning and adaptation in the receiving entity through various methods such as teaching training or data exchange. Inductive Transfer Skill Acquisition Adaptation @@ -3764,8 +4060,8 @@ - The process by which knowledge is passed from one entity such as a person organization or system to another facilitating learning and adaptation in the receiving entity through various methods such as teaching training or data exchange. - https://doi.org/10.1016/j.knosys.2015.01.010 + A training strategy in which knowledge is passed from one entity such as a person organization or system to another facilitating learning and adaptation in the receiving entity through various methods such as teaching training or data exchange. + @@ -3774,6 +4070,18 @@ + + + + + + + + + + + + A network that is an unsupervised technique producing a low-dimensional representation of high-dimensional data preserving topological structure. KN SOFM @@ -3788,7 +4096,7 @@ A network that is an unsupervised technique producing a low-dimensional representation of high-dimensional data preserving topological structure. - https://en.wikipedia.org/wiki/Self-organizing_map + @@ -3799,7 +4107,6 @@ A pooling layer that applies 1D power-average pooling over an input signal composed of several input planes. LPPool1D - LPPool1d LPPool1D Layer @@ -3807,7 +4114,7 @@ A pooling layer that applies 1D power-average pooling over an input signal composed of several input planes. - https://pytorch.org/docs/stable/nn.html#pooling-layers + @@ -3818,7 +4125,6 @@ A pooling layer that applies 2D power-average pooling over an input signal composed of several input planes. LPPool2D - LPPool2d LPPool2D Layer @@ -3826,7 +4132,7 @@ A pooling layer that applies 2D power-average pooling over an input signal composed of several input planes. - https://pytorch.org/docs/stable/nn.html#pooling-layers + @@ -3837,13 +4143,14 @@ A layer that processes one step within the whole time sequence input for an LSTM layer. + Cell class for the LSTM layer. LSTMCell Layer A layer that processes one step within the whole time sequence input for an LSTM layer. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/LSTMCell + @@ -3854,13 +4161,14 @@ A recurrent layer that implements the Long Short-Term Memory architecture. + Long Short-Term Memory layer - Hochreiter 1997. Based on available runtime hardware and constraints, this layer will choose different implementations (cuDNN-based or pure-TensorFlow) to maximize the performance. If a GPU is available and all the arguments to the layer meet the requirement of the cuDNN kernel (see below for details), the layer will use a fast cuDNN implementation. The requirements to use the cuDNN implementation are: 1. activation == tanh, 2. recurrent_activation == sigmoid, 3. recurrent_dropout == 0, 4. unroll is False, 5. use_bias is True, 6. Inputs, if use masking, are strictly right-padded, 7. Eager execution is enabled in the outermost context. LSTM Layer A recurrent layer that implements the Long Short-Term Memory architecture. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/LSTM + @@ -3871,13 +4179,14 @@ A layer that wraps arbitrary expressions as a Layer object. + Wraps arbitrary expressions as a Layer object. Lambda Layer A layer that wraps arbitrary expressions as a Layer object. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Lambda + @@ -3887,7 +4196,6 @@ A large language model that supports interactive semantic parsing enabling users to provide feedback and corrections to dynamically refine and update the language model. - Language Interface LLM Interactive learning Language Interface LLM @@ -3900,7 +4208,6 @@ A model designed to predict the next word in a sequence or assign probabilities to sequences of words in natural language. - Language Model Language Model @@ -3908,7 +4215,7 @@ A model designed to predict the next word in a sequence or assign probabilities to sequences of words in natural language. - https://en.wikipedia.org/wiki/Language_model + @@ -3926,7 +4233,7 @@ A language model consisting of a neural network with many parameters (typically billions of weights or more) trained on large quantities of unlabeled text using self-supervised learning or semi-supervised learning. - https://en.wikipedia.org/wiki/Large_language_model + @@ -3935,15 +4242,15 @@ - A regression analysis method that performs both variable selection and regularization to enhance prediction accuracy and interpretability. + A regression analysis that performs both variable selection and regularization to enhance prediction accuracy and interpretability. Lasso Regression - A regression analysis method that performs both variable selection and regularization to enhance prediction accuracy and interpretability. - https://en.wikipedia.org/wiki/Lasso_(statistics) + A regression analysis that performs both variable selection and regularization to enhance prediction accuracy and interpretability. + @@ -3960,7 +4267,7 @@ A structure or network topology in a deep learning model that takes information from previous layers and passes it to the next layer. - https://en.wikipedia.org/wiki/Layer_(deep_learning) + @@ -3969,15 +4276,16 @@ - The base class from which all layers inherit. + A layer that form which other layers can inherit. + This is the class from which all layers inherit. A layer is a callable object that takes as input one or more tensors and that outputs one or more tensors. It involves computation, defined in the call() method, and a state (weight variables). State can be created in various places, at the convenience of the subclass implementer: in __init__(); in the optional build() method, which is invoked by the first __call__() to the layer, and supplies the shape(s) of the input(s), which may not have been known at initialization time; in the first invocation of call(), with some caveats discussed below. Users will just instantiate a layer and then treat it as a callable. Layer Layer - The base class from which all layers inherit. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Layer + A layer that form which other layers can inherit. + @@ -3989,13 +4297,14 @@ A normalization layer that applies Layer Normalization over a mini-batch of inputs. LayerNorm + Applies Layer Normalization over a mini-batch of inputs as described in the paper Layer Normalization LayerNorm Layer A normalization layer that applies Layer Normalization over a mini-batch of inputs. - https://pytorch.org/docs/stable/nn.html#normalization-layers + @@ -4006,13 +4315,14 @@ A normalization layer that applies Layer Normalization over the inputs. + Layer normalization layer (Ba et al., 2016). Normalize the activations of the previous layer for each given example in a batch independently, rather than across a batch like Batch Normalization. i.e. applies a transformation that maintains the mean activation within each example close to 0 and the activation standard deviation close to 1. Given a tensor inputs, moments are calculated and normalization is performed across the axes specified in axis. LayerNormalization Layer A normalization layer that applies Layer Normalization over the inputs. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/LayerNormalization + @@ -4023,15 +4333,15 @@ A batch normalization layer that lazily initializes the num_features argument from the input size for 1D data. LazyBatchNorm1D - LazyBatchNorm1d + A torch.nn.BatchNorm1D module with lazy initialization of the num_features argument of the BatchNorm1D that is inferred from the input.size(1). LazyBatchNorm1D Layer A batch normalization layer that lazily initializes the num_features argument from the input size for 1D data. - https://pytorch.org/docs/stable/nn.html#normalization-layers + @@ -4042,15 +4352,15 @@ A batch normalization layer that lazily initializes the num_features argument from the input size for 2D data. LazyBatchNorm2D - LazyBatchNorm2d + A torch.nn.BatchNorm2D module with lazy initialization of the num_features argument of the BatchNorm2D that is inferred from the input.size(1). LazyBatchNorm2D Layer A batch normalization layer that lazily initializes the num_features argument from the input size for 2D data. - https://pytorch.org/docs/stable/nn.html#normalization-layers + @@ -4061,72 +4371,75 @@ A batch normalization layer that lazily initializes the num_features argument from the input size for 3D data. LazyBatchNorm3D - LazyBatchNorm3d + A torch.nn.BatchNorm3D module with lazy initialization of the num_features argument of the BatchNorm3D that is inferred from the input.size(1). LazyBatchNorm3D Layer A batch normalization layer that lazily initializes the num_features argument from the input size for 3D data. - https://pytorch.org/docs/stable/nn.html#normalization-layers + - + - + - An instance normalization layer that lazily initializes the num_features argument from the input size for 1D data. + A normalization layer that lazily initializes the num_features argument from the input size for 1D data. LazyInstanceNorm1D - LazyInstanceNorm1d + - LazyInstanceNorm1d Layer + A torch.nn.InstanceNorm1D module with lazy initialization of the num_features argument of the InstanceNorm1D that is inferred from the input.size(1). + LazyInstanceNorm1D Layer - + - An instance normalization layer that lazily initializes the num_features argument from the input size for 1D data. - https://pytorch.org/docs/stable/nn.html#normalization-layers + A normalization layer that lazily initializes the num_features argument from the input size for 1D data. + - + - + - An instance normalization layer that lazily initializes the num_features argument from the input size for 2D data. + A normalization layer that lazily initializes the num_features argument from the input size for 2D data. LazyInstanceNorm2D - LazyInstanceNorm2d + - LazyInstanceNorm2d Layer + A torch.nn.InstanceNorm2D module with lazy initialization of the num_features argument of the InstanceNorm2D that is inferred from the input.size(1). + LazyInstanceNorm2D Layer - + - An instance normalization layer that lazily initializes the num_features argument from the input size for 2D data. - https://pytorch.org/docs/stable/nn.html#normalization-layers + A normalization layer that lazily initializes the num_features argument from the input size for 2D data. + - + - + - An instance normalization layer that lazily initializes the num_features argument from the input size for 3D data. + A normalization that lazily initializes the num_features argument from the input size for 3D data. LazyInstanceNorm3D - LazyInstanceNorm3d + - LazyInstanceNorm3d Layer + A torch.nn.InstanceNorm3D module with lazy initialization of the num_features argument of the InstanceNorm3D that is inferred from the input.size(1). + LazyInstanceNorm3D Layer - + - An instance normalization layer that lazily initializes the num_features argument from the input size for 3D data. - https://pytorch.org/docs/stable/nn.html#normalization-layers + A normalization that lazily initializes the num_features argument from the input size for 3D data. + @@ -4137,13 +4450,14 @@ An activation layer that applies the leaky rectified linear unit function element-wise. + Leaky version of a Rectified Linear Unit. LeakyReLU Layer An activation layer that applies the leaky rectified linear unit function element-wise. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/LeakyReLU + @@ -4160,7 +4474,7 @@ A regression analysis which approximates the solution of overdetermined systems by minimizing the sum of the squares of the residuals. - https://en.wikipedia.org/wiki/Least_squares + @@ -4172,7 +4486,6 @@ A large language model that continually acquires new knowledge over time without forgetting previously learned information maintaining a balance between plasticity and stability. Continual Learning LLM Forever Learning - Lifelong Learning LLM Catastrophic forgetting Plasticity-Stability balance @@ -4184,16 +4497,17 @@ - - An activation function that has the form f(x) = a + bx. + + A mathematical function that has the form f(x) = a + bx. + Linear Function - An activation function that has the form f(x) = a + bx. - https://www.tensorflow.org/api_docs/python/tf/keras/activations/linear + A mathematical function that has the form f(x) = a + bx. + @@ -4202,15 +4516,15 @@ - A regression analysis model that is a linear approach for modeling the relationship between a scalar response and one or more explanatory variables. + A regression analysis that is a linear approach for modeling the relationship between a scalar response and one or more explanatory variables. Linear Regression - A regression analysis model that is a linear approach for modeling the relationship between a scalar response and one or more explanatory variables. - https://en.wikipedia.org/wiki/Linear_regression + A regression analysis that is a linear approach for modeling the relationship between a scalar response and one or more explanatory variables. + @@ -4221,14 +4535,14 @@ A use and interpretation bias arising when network attributes obtained from user connections activities or interactions misrepresent true user behavior. + Bias arising when network attributes obtained from user connections, activities, or interactions misrepresent true user behavior. Linking Bias A use and interpretation bias arising when network attributes obtained from user connections activities or interactions misrepresent true user behavior. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -4237,6 +4551,24 @@ + + + + + + + + + + + + + + + + + + A network that is a type of reservoir computer turning time-varying input into spatio-temporal activation patterns. LSM @@ -4247,7 +4579,7 @@ A network that is a type of reservoir computer turning time-varying input into spatio-temporal activation patterns. - https://en.wikipedia.org/wiki/Liquid_state_machine + @@ -4259,13 +4591,14 @@ A normalization layer that applies local response normalization over an input signal composed of several input planes. LocalResponseNorm + Applies local response normalization over an input signal composed of several input planes, where channels occupy the second dimension. LocalResponseNorm Layer A normalization layer that applies local response normalization over an input signal composed of several input planes. - https://pytorch.org/docs/stable/nn.html#normalization-layers + @@ -4276,13 +4609,14 @@ A locally-connected layer for 1D inputs where each patch of the input is convolved with a different set of filters. + Locally-connected layer for 1D inputs. The LocallyConnected1D layer works similarly to the Conv1D layer, except that weights are unshared, that is, a different set of filters is applied at each different patch of the input. LocallyConnected1D Layer A locally-connected layer for 1D inputs where each patch of the input is convolved with a different set of filters. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/LocallyConnected1D + @@ -4293,13 +4627,14 @@ A locally-connected layer for 2D inputs where each patch of the input is convolved with a different set of filters. + Locally-connected layer for 2D inputs. The LocallyConnected2D layer works similarly to the Conv2D layer, except that weights are unshared, that is, a different set of filters is applied at each different patch of the input. LocallyConnected2D Layer A locally-connected layer for 2D inputs where each patch of the input is convolved with a different set of filters. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/LocallyConnected2D + @@ -4316,7 +4651,7 @@ A layer that works similarly to the Convolution1D layer, except that weights are unshared, that is, a different set of filters is applied at each different patch of the input. - https://faroit.com/keras-docs/1.2.2/layers/local/ + @@ -4325,15 +4660,15 @@ - A regression analysis model that estimates the probability of an event occurring by modeling the log-odds of the event as a linear combination of one or more independent variables. + A regression analysis that estimates the probability of an event occurring by modeling the log-odds of the event as a linear combination of one or more independent variables. Logistic Regression - A regression analysis model that estimates the probability of an event occurring by modeling the log-odds of the event as a linear combination of one or more independent variables. - https://en.wikipedia.org/wiki/Logistic_regression + A regression analysis that estimates the probability of an event occurring by modeling the log-odds of the event as a linear combination of one or more independent variables. + @@ -4342,6 +4677,24 @@ + + + + + + + + + + + + + + + + + + A recurrent neural network with feedback connections that processes entire sequences of data. LSTM @@ -4352,7 +4705,7 @@ A recurrent neural network with feedback connections that processes entire sequences of data. - https://en.wikipedia.org/wiki/Long_short-term_memory + @@ -4369,8 +4722,7 @@ An individual bias occurring when automation leads to humans being unaware of their situation making them unprepared to assume control in cooperative systems. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -4379,7 +4731,7 @@ - A LLM which is optimized for performance in scenarios with limited data, computational resources, or for languages with sparse datasets. + A large language model which is optimized for performance in scenarios with limited data, computational resources, or for languages with sparse datasets. Low-Resource Language Model low-resource languages resource-efficient @@ -4389,19 +4741,20 @@ - + - + A field of inquiry devoted to understanding and building methods that learn from data to improve performance on a set of tasks. - Machine Learning + Machine Learning + + Machine Learning Task - + A field of inquiry devoted to understanding and building methods that learn from data to improve performance on a set of tasks. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -4418,7 +4771,7 @@ A dimensionality reduction method based on the assumption that observed data lie on a low-dimensional manifold embedded in a higher-dimensional space. - https://arxiv.org/abs/2011.01307 + @@ -4427,6 +4780,12 @@ + + + + + + A network that is a stochastic model describing a sequence of possible events where the probability of each event depends only on the previous event's state. MC MP @@ -4439,7 +4798,7 @@ A network that is a stochastic model describing a sequence of possible events where the probability of each event depends only on the previous event's state. - https://en.wikipedia.org/wiki/Markov_chain + @@ -4449,7 +4808,6 @@ A language model that is trained to predict randomly masked tokens in a sequence based on the remaining unmasked tokens allowing it to build deep bidirectional representations that can be effectively transferred to various NLP tasks via fine-tuning. - Masked Language Model bidirectional encoder denoising autoencoder @@ -4464,13 +4822,14 @@ A layer that masks a sequence by using a mask value to skip timesteps. + Masks a sequence by using a mask value to skip timesteps. For each timestep in the input tensor (dimension #1 in the tensor), if all values in the input tensor at that timestep are equal to mask_value, then the timestep will be masked (skipped) in all downstream layers (as long as they support masking). If any downstream layer does not support masking yet receives such an input mask, an exception will be raised. Masking Layer A layer that masks a sequence by using a mask value to skip timesteps. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Masking + @@ -4479,30 +4838,43 @@ - An input layer with a shape corresponding to that of the output layer. - - Matched Input-Output Layer + + + + + A mathematical rule that gives the value of a dependent variable corresponding to specified values of one or more independent variables. + + Mathematical Function + + + + + A mathematical rule that gives the value of a dependent variable corresponding to specified values of one or more independent variables. + + + + + A pooling layer that performs max pooling operation for temporal data. MaxPool1D - MaxPool1d MaxPooling1D - MaxPooling1d + Max pooling operation for 1D temporal data. Downsamples the input representation by taking the maximum value over a spatial window of size pool_size. The window is shifted by strides. The resulting output, when using the "valid" padding option, has a shape of: output_shape = (input_shape - pool_size + 1) / strides) The resulting output shape when using the "same" padding option is: output_shape = input_shape / strides. MaxPooling1D Layer A pooling layer that performs max pooling operation for temporal data. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/MaxPool1D + @@ -4513,17 +4885,16 @@ A pooling layer that performs max pooling operation for spatial data. MaxPool2D - MaxPool2d MaxPooling2D - MaxPooling2d + Max pooling operation for 2D spatial data. MaxPooling2D Layer A pooling layer that performs max pooling operation for spatial data. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/MaxPool2D + @@ -4534,17 +4905,16 @@ A pooling layer that performs max pooling operation for 3D data (spatial or spatio-temporal). MaxPool3D - MaxPool3d MaxPooling3D - MaxPooling3d + Max pooling operation for 3D data (spatial or spatio-temporal). Downsamples the input along its spatial dimensions (depth, height, and width) by taking the maximum value over an input window (of size defined by pool_size) for each channel of the input. The window is shifted by strides along each dimension. MaxPooling3D Layer A pooling layer that performs max pooling operation for 3D data (spatial or spatio-temporal). - https://www.tensorflow.org/api_docs/python/tf/keras/layers/MaxPool3D + @@ -4553,17 +4923,17 @@ - A pooling layer that computes a partial inverse of MaxPool1d. + A pooling layer that computes a partial inverse of MaxPool1D. MaxUnpool1D - MaxUnpool1d + Computes a partial inverse of MaxPool1D. MaxUnpool1D Layer - A pooling layer that computes a partial inverse of MaxPool1d. - https://pytorch.org/docs/stable/nn.html#pooling-layers + A pooling layer that computes a partial inverse of MaxPool1D. + @@ -4572,17 +4942,17 @@ - A pooling layer that computes a partial inverse of MaxPool2d. + A pooling layer that computes a partial inverse of MaxPool2D. MaxUnpool2D - MaxUnpool2d + Computes a partial inverse of MaxPool2D. MaxUnpool2D Layer - A pooling layer that computes a partial inverse of MaxPool2d. - https://pytorch.org/docs/stable/nn.html#pooling-layers + A pooling layer that computes a partial inverse of MaxPool2D. + @@ -4591,17 +4961,17 @@ - A pooling layer that computes a partial inverse of MaxPool3d. + A pooling layer that computes a partial inverse of MaxPool3D. MaxUnpool3D - MaxUnpool3d + Computes a partial inverse of MaxPool3D. MaxUnpool3D Layer - A pooling layer that computes a partial inverse of MaxPool3d. - https://pytorch.org/docs/stable/nn.html#pooling-layers + A pooling layer that computes a partial inverse of MaxPool3D. + @@ -4612,13 +4982,14 @@ A merging layer that computes the maximum (element-wise) of a list of inputs. + Layer that computes the maximum (element-wise) a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape). Maximum Layer A merging layer that computes the maximum (element-wise) of a list of inputs. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Maximum + @@ -4635,8 +5006,7 @@ A selection and sampling bias arising when features and labels are proxies for desired quantities potentially leading to differential performance. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -4645,7 +5015,7 @@ - A LLM which incorporates external writable and readable memory components, allowing it to store and retrieve information over long contexts. + A large language model which incorporates external writable and readable memory components, allowing it to store and retrieve information over long contexts. Memory-Augmented Large Language Model external memory @@ -4654,8 +5024,8 @@ - A LLM which incorporates external writable and readable memory components, allowing it to store and retrieve information over long contexts. - https://arxiv.org/abs/2306.07174 + A large language model which incorporates external writable and readable memory components, allowing it to store and retrieve information over long contexts. + @@ -4664,16 +5034,7 @@ - A layer of cells, each with an internal state or weights. - - Memory Cell Layer - - - - A layer of cells, each with an internal state or weights. - https://doi.org/10.1162/neco.1997.9.8.1735 - @@ -4689,7 +5050,7 @@ A layer used to merge a list of inputs. - https://www.tutorialspoint.com/keras/keras_merge_layer.htm + @@ -4697,17 +5058,16 @@ - - A machine learning that automatically learns from metadata about machine learning experiments. + + A machine learning task that automatically learns from metadata about machine learning experiments. Meta-Learning - A machine learning that automatically learns from metadata about machine learning experiments. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + A machine learning task that automatically learns from metadata about machine learning experiments. + @@ -4716,7 +5076,7 @@ - A LLM which is trained in a way that allows it to quickly adapt to new tasks or datasets through only a few examples or fine-tuning steps, leveraging meta-learned priors about how to efficiently learn. + A large language model which is trained in a way that allows it to quickly adapt to new tasks or datasets through only a few examples or fine-tuning steps, leveraging meta-learned priors about how to efficiently learn. Meta-Learning Large Language Model few-shot adaptation learning to learn @@ -4733,13 +5093,14 @@ A deep neural network that learns a representation function mapping objects into an embedded space. Distance Metric Learning + Learning a representation function that maps objects into an embedded space. Metric Learning A deep neural network that learns a representation function mapping objects into an embedded space. - https://paperswithcode.com/task/metric-learning + @@ -4750,13 +5111,14 @@ A merging layer that computes the minimum (element-wise) of a list of inputs. + Layer that computes the minimum (element-wise) a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape). Minimum Layer A merging layer that computes the minimum (element-wise) of a list of inputs. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Minimum + @@ -4765,7 +5127,7 @@ - A LLM which dynamically selects and combines outputs from multiple expert submodels, allowing for efficient scaling by conditionally activating only a subset of model components for each input. + A large language model which dynamically selects and combines outputs from multiple expert submodels, allowing for efficient scaling by conditionally activating only a subset of model components for each input. Mixture-of-Experts Large Language Model MoE Large Language Model conditional computation @@ -4776,8 +5138,8 @@ - A LLM which dynamically selects and combines outputs from multiple expert submodels, allowing for efficient scaling by conditionally activating only a subset of model components for each input. - https://proceedings.mlr.press/v162/du22c.html + A large language model which dynamically selects and combines outputs from multiple expert submodels, allowing for efficient scaling by conditionally activating only a subset of model components for each input. + @@ -4786,16 +5148,15 @@ - A bias occurring when modal interfaces confuse human operators causing actions appropriate for a different mode but incorrect for the current situation. + An individual bias occurring when modal interfaces confuse human operators causing actions appropriate for a different mode but incorrect for the current situation. Mode Confusion Bias - A bias occurring when modal interfaces confuse human operators causing actions appropriate for a different mode but incorrect for the current situation. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + An individual bias occurring when modal interfaces confuse human operators causing actions appropriate for a different mode but incorrect for the current situation. + @@ -4805,7 +5166,6 @@ An abstract representation of a complex system generally assembled as a set of logical mathematical or conceptual properties to simulate or understand the system's behavior. - Model Model @@ -4813,7 +5173,7 @@ An abstract representation of a complex system generally assembled as a set of logical mathematical or conceptual properties to simulate or understand the system's behavior. - https://en.wikipedia.org/wiki/Mathematical_model + @@ -4822,7 +5182,7 @@ - Techniques aimed at making models more efficient such as knowledge distillation. + A preprocessing aimed at making models more efficient, such as knowledge distillation. Computational Efficiency Model Optimization @@ -4831,8 +5191,8 @@ - Techniques aimed at making models more efficient such as knowledge distillation. - https://doi.org/10.1145/3578938 + A preprocessing aimed at making models more efficient, such as knowledge distillation. + @@ -4849,8 +5209,7 @@ A processing bias introduced when using data to select a single "best" model from many or when an explanatory variable has a weak relationship with the response variable. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -4859,7 +5218,7 @@ - A modular large language model that consists of multiple specialized components or skills that can be dynamically composed and recombined to solve complex tasks mimicking the modular structure of human cognition. + A modular language model that consists of multiple specialized components or skills that can be dynamically composed and recombined to solve complex tasks mimicking the modular structure of human cognition. Modular Large Language Model component skills skill composition @@ -4869,8 +5228,8 @@ - A modular large language model that consists of multiple specialized components or skills that can be dynamically composed and recombined to solve complex tasks mimicking the modular structure of human cognition. - https://arxiv.org/abs/2302.11529v2 + A modular language model that consists of multiple specialized components or skills that can be dynamically composed and recombined to solve complex tasks mimicking the modular structure of human cognition. + @@ -4888,7 +5247,7 @@ A language model that consists of multiple specialized components or skills that can be dynamically composed and recombined to solve complex tasks mimicking the modular structure of human cognition. - https://arxiv.org/abs/2302.11529v2 + @@ -4899,13 +5258,14 @@ An attention layer that allows the model to attend to information from different representation subspaces. + MultiHeadAttention layer. This is an implementation of multi-headed attention as described in the paper "Attention is all you Need" (Vaswani et al., 2017). If query, key, value are the same, then this is self-attention. Each timestep in query attends to the corresponding sequence in key, and returns a fixed-width vector.This layer first projects query, key and value. These are (effectively) a list of tensors of length num_attention_heads, where the corresponding shapes are (batch_size, <query dimensions>, key_dim), (batch_size, <key/value dimensions>, key_dim), (batch_size, <key/value dimensions>, value_dim).Then, the query and key tensors are dot-producted and scaled. These are softmaxed to obtain attention probabilities. The value tensors are then interpolated by these probabilities, then concatenated back to a single tensor. Finally, the result tensor with the last dimension as value_dim can take an linear projection and return. When using MultiHeadAttention inside a custom Layer, the custom Layer must implement build() and call MultiHeadAttention's _build_from_signature(). This enables weights to be restored correctly when the model is loaded. MultiHeadAttention Layer An attention layer that allows the model to attend to information from different representation subspaces. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/MultiHeadAttention + @@ -4914,7 +5274,7 @@ - A LLM which is trained jointly on multiple language tasks simultaneously, learning shared representations that transfer across tasks. + A large language model which is trained jointly on multiple language tasks simultaneously, learning shared representations that transfer across tasks. Multi-Task Large Language Model transfer learning @@ -4927,16 +5287,17 @@ - A machine learning task focused on methods that classify instances into one of three or more classes. + A classification focused on methods that classify instances into one of three or more classes. Multinomial Classification + Methods that classify instances into one of three or more classes. Multiclass Classification - A machine learning task focused on methods that classify instances into one of three or more classes. - https://en.wikipedia.org/wiki/Multiclass_classification + A classification focused on methods that classify instances into one of three or more classes. + @@ -4954,7 +5315,7 @@ A dimensionality reduction method that translates information about the pairwise distances among a set of objects or individuals into a configuration of points mapped into an abstract Cartesian space. - https://en.wikipedia.org/wiki/Multidimensional_scaling + @@ -4978,13 +5339,14 @@ A deep neural network that processes and links information using various modalities. + Creating models that process and link information using various modalities. Multimodal Deep Learning A deep neural network that processes and links information using various modalities. - https://arxiv.org/abs/2105.11087 + @@ -4994,7 +5356,6 @@ A large language model that learns joint representations across different modalities like text vision and audio in an end-to-end fashion for better cross-modal understanding and generation. - Multimodal Fusion LLM cross-modal grounding Multimodal Fusion LLM @@ -5016,7 +5377,7 @@ A multimodal large language model that learns joint representations across different modalities like text vision and audio in an end-to-end fashion for better cross-modal understanding and generation. - https://arxiv.org/abs/2303.17580 + @@ -5034,7 +5395,7 @@ A language model that learns joint representations across different modalities like text vision and audio in an end-to-end fashion for better cross-modal understanding and generation. - https://arxiv.org/abs/2205.12630 + @@ -5042,17 +5403,17 @@ - - A type of machine learning that uses multiple modalities of data such as text audio and images to improve learning outcomes. + + A machine learning task that uses multiple modalities of data such as text audio and images to improve learning outcomes. + A type of deep learning that uses multiple modalities of data, such as text, audio, and images, to improve learning outcomes. Multimodal Learning - A type of machine learning that uses multiple modalities of data such as text audio and images to improve learning outcomes. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + A machine learning task that uses multiple modalities of data such as text audio and images to improve learning outcomes. + @@ -5061,16 +5422,15 @@ - A multimodal LLM which processes prompts that include multiple modalities, such as both text and images, to generate relevant responses. - Multimodal Prompt-based Language Model + A multimodal large language model which processes prompts that include multiple modalities, such as both text and images, to generate relevant responses. Multimodal Prompt-based Language Model - A multimodal LLM which processes prompts that include multiple modalities, such as both text and images, to generate relevant responses. - https://arxiv.org/abs/2210.03094 + A multimodal large language model which processes prompts that include multiple modalities, such as both text and images, to generate relevant responses. + @@ -5080,7 +5440,6 @@ A transformer network that processes and relates information from different modalities such as text images and audio using a shared embedding space and attention mechanism to learn joint representations across modalities. - Multimodal Transformer unified encoder vision-language model @@ -5095,13 +5454,14 @@ A merging layer that multiplies (element-wise) a list of inputs. + Layer that multiplies (element-wise) a list of inputs. It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape). Multiply Layer A merging layer that multiplies (element-wise) a list of inputs. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Multiply + @@ -5109,8 +5469,8 @@ - - A subfield of machine learning focused on the interactions between computers and human language including programming computers to process and analyze large amounts of natural language data. + + A machine learning task focused on the interactions between computers and human language including programming computers to process and analyze large amounts of natural language data. NLP Natural Language Processing @@ -5118,8 +5478,8 @@ - A subfield of machine learning focused on the interactions between computers and human language including programming computers to process and analyze large amounts of natural language data. - https://en.wikipedia.org/wiki/Natural_language_processing + A machine learning task focused on the interactions between computers and human language including programming computers to process and analyze large amounts of natural language data. + @@ -5139,7 +5499,30 @@ - + + + + + + + + + + + + + + + + + + + + + + + + A deep feedforward network that combines neural network pattern matching with the algorithmic power of programmable computers. NTM @@ -5150,7 +5533,7 @@ A deep feedforward network that combines neural network pattern matching with the algorithmic power of programmable computers. - https://en.wikipedia.org/wiki/Neural_Turing_machine + @@ -5159,7 +5542,7 @@ - A LLM which combines neural language modeling with symbolic reasoning components, leveraging structured knowledge representations and logical inferences to improve reasoning capabilities. + A large language model which combines neural language modeling with symbolic reasoning components, leveraging structured knowledge representations and logical inferences to improve reasoning capabilities. Neuro-Symbolic Large Language Model knowledge reasoning symbolic grounding @@ -5175,13 +5558,14 @@ A layer that is a densely-connected neural network layer with added noise for regularization. + Noisy dense layer that injects random noise to the weights of dense layer. Noisy dense layers are fully connected layers whose weights and biases are augmented by factorised Gaussian noise. The factorised Gaussian noise is controlled through gradient descent by a second weights layer. A NoisyDense layer implements the operation: $$ mathrm{NoisyDense}(x) = mathrm{activation}(mathrm{dot}(x, mu + (sigma cdot epsilon)) mathrm{bias}) $$ where mu is the standard weights layer, epsilon is the factorised Gaussian noise, and delta is a second weights layer which controls epsilon. Noise Dense Layer A layer that is a densely-connected neural network layer with added noise for regularization. - https://www.tensorflow.org/addons/api_docs/python/tfa/layers/NoisyDense + @@ -5190,16 +5574,7 @@ - An input layer that adds noise to each value. - - Noisy Input Layer - - - - An input layer that adds noise to each value. - https://doi.org/10.1109/21.155944 - @@ -5207,7 +5582,7 @@ - The technique of transforming data into a standard format or scale typically to reduce redundancy and improve consistency often involving the adjustment of values measured on different scales to a common scale. + A data preparation that transforms data into a standard format or scale typically to reduce redundancy and improve consistency often involving the adjustment of values measured on different scales to a common scale. Normalization @@ -5218,15 +5593,15 @@ - A preprocessing layer that normalizes continuous features. + A numerical features prepreprocessing layer that normalizes continuous features. Normalization Layer - A preprocessing layer that normalizes continuous features. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Normalization + A numerical features prepreprocessing layer that normalizes continuous features. + @@ -5243,7 +5618,7 @@ A layer that performs numerical data preprocessing operations. - https://keras.io/guides/preprocessing_layers/ + @@ -5261,7 +5636,7 @@ A deep neural network that classified objects from one or only a few examples. - https://en.wikipedia.org/wiki/One-shot_learning + @@ -5284,16 +5659,7 @@ - A layer containing the last neurons in the network that produces given outputs for the program. - - Output Layer - - - - A layer containing the last neurons in the network that produces given outputs for the program. - https://www.techopedia.com/definition/33263/output-layer-neural-networks - @@ -5303,13 +5669,14 @@ An activation layer that applies parametric rectified linear unit function element-wise. + Parametric Rectified Linear Unit. PReLU Layer An activation layer that applies parametric rectified linear unit function element-wise. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/PReLU + @@ -5318,6 +5685,18 @@ + + + + + + + + + + + + An artificial neural network with a supervised learning algorithm for binary classification using a linear predictor function. FFN Feed-Forward Network @@ -5334,15 +5713,16 @@ - A layer that permutes the dimensions of the input according to a given pattern. + A reshaping layer that permutes the dimensions of the input according to a given pattern. + Permutes the dimensions of the input according to a given pattern. Useful e.g. connecting RNNs and convnets. Permute Layer - A layer that permutes the dimensions of the input according to a given pattern. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Permute + A reshaping layer that permutes the dimensions of the input according to a given pattern. + @@ -5364,9 +5744,6 @@ - A layer that, after taking a set of states or values as input, predicts a probability distribution of actions to take. - - Policy Layer @@ -5375,16 +5752,7 @@ - A layer that serves to mitigate the sensitivity of convolutional layers to location and spatially downsample representations. - - Pooling Layer - - - - A layer that serves to mitigate the sensitivity of convolutional layers to location and spatially downsample representations. - https://d2l.ai/chapter_convolutional-neural-networks/pooling.html - @@ -5394,14 +5762,14 @@ A selection and sampling bias where more popular items are more exposed under-representing less popular items. + Selection bias where more popular items are more exposed, under-representing less popular items. Popularity Bias A selection and sampling bias where more popular items are more exposed under-representing less popular items. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -5418,8 +5786,7 @@ A selection and sampling bias characterized by systematic distortions in demographics or other user characteristics between represented users and the target population. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -5436,7 +5803,7 @@ The series of steps applied to raw data before it is used in a machine learning model including tasks such as normalization scaling encoding and transformation to ensure the data is in an appropriate format and quality for analysis. - https://doi.org/10.1109/ICDE.2019.00245 + @@ -5453,7 +5820,7 @@ A layer that performs data preprocessing operations. - https://www.tensorflow.org/guide/keras/preprocessing_layers + @@ -5470,8 +5837,7 @@ An individual bias arising from how information is presented on the Web via a user interface due to rating or ranking of output or through users' self-selected biased interaction. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -5489,7 +5855,7 @@ A dimensionality reduction method for analyzing large datasets with high-dimensional features per observation increasing data interpretability while preserving maximum information and enabling visualization. - https://en.wikipedia.org/wiki/Principal_component_analysis + @@ -5497,8 +5863,8 @@ - - A machine learning model in which a graph expresses the conditional dependence structure between random variables. + + A machine learning task in which a graph expresses the conditional dependence structure between random variables. Graphical Model PGM Structure Probabilistic Model @@ -5508,8 +5874,8 @@ - A machine learning model in which a graph expresses the conditional dependence structure between random variables. - https://en.wikipedia.org/wiki/Graphical_model + A machine learning task in which a graph expresses the conditional dependence structure between random variables. + @@ -5518,9 +5884,6 @@ - A hidden layer that estimates the probability of a sample being within a certain category. - - Probabilistic Hidden Layer @@ -5537,7 +5900,7 @@ A probabilistic graphical model that uses statistical techniques to analyze the words in each text to discover common themes their connections and their changes over time. - https://pyro.ai/examples/prodlda.html + @@ -5549,14 +5912,14 @@ A computational bias resulting from judgment modulated by affect influenced by the level of efficacy and efficiency in information processing. Validation Bias + Judgment modulated by affect, influenced by the level of efficacy and efficiency in information processing; often referred to as aesthetic judgment in cognitive sciences. Processing Bias A computational bias resulting from judgment modulated by affect influenced by the level of efficacy and efficiency in information processing. - GTP-4o with Seppala et al. 2017 - https://en.wikipedia.org/wiki/Bias_(statistics) + @@ -5565,7 +5928,7 @@ - A LLM which is fine-tuned on a small number of examples or prompts, rather than full task datasets. This allows for rapid adaptation to new tasks with limited data, leveraging the model's few-shot learning capabilities. + A large language model which is fine-tuned on a small number of examples or prompts, rather than full task datasets. This allows for rapid adaptation to new tasks with limited data, leveraging the model's few-shot learning capabilities. Prompt-based Fine-Tuning Large Language Model Prompt-tuned Large Language Model few-shot learning @@ -5579,17 +5942,16 @@ - - A regression analysis method for survival analysis where the unique effect of a unit increase in a covariate is multiplicative with respect to the hazard rate. + A regression analysis for survival analysis where the unique effect of a unit increase in a covariate is multiplicative with respect to the hazard rate. Proportional Hazards Model - A regression analysis method for survival analysis where the unique effect of a unit increase in a covariate is multiplicative with respect to the hazard rate. - https://en.wikipedia.org/wiki/Proportional_hazards_model + A regression analysis for survival analysis where the unique effect of a unit increase in a covariate is multiplicative with respect to the hazard rate. + @@ -5600,13 +5962,14 @@ The base class for recurrent layers. + Base class for recurrent layers. RNN Layer The base class for recurrent layers. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/RNN + @@ -5615,6 +5978,24 @@ + + + + + + + + + + + + + + + + + + A deep feedforward network that uses radial basis functions as activation functions for pattern recognition and interpolation. RBFN RBN @@ -5627,7 +6008,7 @@ A deep feedforward network that uses radial basis functions as activation functions for pattern recognition and interpolation. - https://en.wikipedia.org/wiki/Radial_basis_function_network + @@ -5635,16 +6016,17 @@ - + An image preprocessing layer that randomly adjusts brightness during training. + A preprocessing layer which randomly adjusts brightness during training. This layer will randomly increase/reduce the brightness for the input RGB images. At inference time, the output will be identical to the input. Call the layer with training=True to adjust the brightness of the input. Note that different brightness adjustment factors will be apply to each the images in the batch. RandomBrightness Layer An image preprocessing layer that randomly adjusts brightness during training. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomBrightness + @@ -5652,16 +6034,17 @@ - + An image preprocessing layer that randomly adjusts contrast during training. + A preprocessing layer which randomly adjusts contrast during training. This layer will randomly adjust the contrast of an image or images by a random factor. Contrast is adjusted independently for each channel of each image during training. For each channel, this layer computes the mean of the image pixels in the channel and then adjusts each component x of each pixel to (x - mean) * contrast_factor + mean. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and in integer or floating point dtype. By default, the layer will output floats. The output value will be clipped to the range [0, 255], the valid range of RGB colors. RandomContrast Layer An image preprocessing layer that randomly adjusts contrast during training. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomContrast + @@ -5669,16 +6052,17 @@ - + An image preprocessing layer that randomly crops images during training. + A preprocessing layer which randomly crops images during training. During training, this layer will randomly choose a location to crop images down to a target size. The layer will crop all the images in the same batch to the same cropping location. At inference time, and during training if an input image is smaller than the target size, the input will be resized and cropped so as to return the largest possible window in the image that matches the target aspect ratio. If you need to apply random cropping at inference time, set training to True when calling the layer. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. RandomCrop Layer An image preprocessing layer that randomly crops images during training. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomCrop + @@ -5687,7 +6071,7 @@ - A regression analysis model where the model parameters are random variables. + A regression analysis where the model parameters are random variables. REM Random Effects Model @@ -5695,8 +6079,8 @@ - A regression analysis model where the model parameters are random variables. - https://en.wikipedia.org/wiki/Random_effects_model + A regression analysis where the model parameters are random variables. + @@ -5704,16 +6088,17 @@ - + An image preprocessing layer that randomly flips images during training. + A preprocessing layer which randomly flips images during training. This layer will flip the images horizontally and or vertically based on the mode attribute. During inference time, the output will be identical to input. Call the layer with training=True to flip the input. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. RandomFlip Layer An image preprocessing layer that randomly flips images during training. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomFlip + @@ -5730,7 +6115,7 @@ An ensemble learning method for classification regression and other tasks that constructs a multitude of decision trees during training. - https://en.wikipedia.org/wiki/Random_forest + @@ -5738,16 +6123,17 @@ - + An image preprocessing layer that randomly varies image height during training. + A preprocessing layer which randomly varies image height during training. This layer adjusts the height of a batch of images by a random factor. The input should be a 3D (unbatched) or 4D (batched) tensor in the "channels_last" image data format. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. By default, this layer is inactive during inference. RandomHeight Layer An image preprocessing layer that randomly varies image height during training. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomHeight + @@ -5755,16 +6141,17 @@ - + An image preprocessing layer that randomly rotates images during training. + A preprocessing layer which randomly rotates images during training. RandomRotation Layer An image preprocessing layer that randomly rotates images during training. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomRotation + @@ -5772,16 +6159,17 @@ - + An image preprocessing layer that randomly translates images during training. + A preprocessing layer which randomly translates images during training. This layer will apply random translations to each image during training, filling empty space according to fill_mode. aInput pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. RandomTranslation Layer An image preprocessing layer that randomly translates images during training. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomTranslation + @@ -5789,16 +6177,17 @@ - + An image preprocessing layer that randomly varies image width during training. + A preprocessing layer which randomly varies image width during training. This layer will randomly adjusts the width of a batch of images of a batch of images by a random factor. The input should be a 3D (unbatched) or 4D (batched) tensor in the "channels_last" image data format. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. By default, this layer is inactive during inference. RandomWidth Layer An image preprocessing layer that randomly varies image width during training. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomWidth + @@ -5806,16 +6195,17 @@ - + An image preprocessing layer that randomly zooms in or out on images during training. + A preprocessing layer which randomly zooms images during training. This layer will randomly zoom in or out on each axis of an image independently, filling empty space according to fill_mode.Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. RandomZoom Layer An image preprocessing layer that randomly zooms in or out on images during training. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/RandomZoom + @@ -5832,8 +6222,7 @@ An anchoring bias characterized by the idea that top-ranked results are the most relevant and important leading to more clicks than other results. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -5846,14 +6235,14 @@ Rashomon Effect Rashomon Principle + Differences in perspective, memory, recall, interpretation, and reporting of the same event by multiple persons or witnesses. Rashomon Effect Bias An individual bias characterized by differences in perspective memory recall interpretation and reporting of the same event by multiple persons or witnesses. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -5861,18 +6250,20 @@ - - An activation function that returns max(x 0) the element-wise maximum of 0 and the input tensor. + + A mathematical function that returns max(x 0) the element-wise maximum of 0 and the input tensor. ReLU Rectified Linear Unit + + The ReLU activation function returns: max(x, 0), the element-wise maximum of 0 and the input tensor. ReLU Function - An activation function that returns max(x 0) the element-wise maximum of 0 and the input tensor. - https://www.tensorflow.org/api_docs/python/tf/keras/activations/relu + A mathematical function that returns max(x 0) the element-wise maximum of 0 and the input tensor. + @@ -5881,16 +6272,7 @@ - An activation layer that applies the rectified linear unit function element-wise. - - ReLU Layer - - - - An activation layer that applies the rectified linear unit function element-wise. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/ReLU - @@ -5910,7 +6292,7 @@ A large language model that incorporates explicit reasoning capabilities leveraging logical rules axioms or external knowledge to make deductive inferences during language tasks. - https://doi.org/10.18653/v1/2023.acl-long.347 + @@ -5919,16 +6301,7 @@ - A layer composed of recurrent units with the number equal to the hidden size of the layer. - - Recurrent Layer - - - - A layer composed of recurrent units with the number equal to the hidden size of the layer. - https://docs.nvidia.com/deepLearning/performance/dl-performance-recurrent/index.html#recurrent-layer - @@ -5936,7 +6309,7 @@ - A deep neural network with connections forming a directed graph along a temporal sequence enabling dynamic behavior. + A network with connections forming a directed graph along a temporal sequence enabling dynamic behavior. RN RecNN Recurrent Network @@ -5950,7 +6323,7 @@ - A large language model that uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions improving systematic generalization abilities. + A language model that uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions improving systematic generalization abilities. Recursive Large Language Model Self-Attending Large Language Model iterative refinement @@ -5961,8 +6334,8 @@ - A large language model that uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions improving systematic generalization abilities. - https://doi.org/10.1609/aaai.v33i01.33017450 + A language model that uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions improving systematic generalization abilities. + @@ -5971,7 +6344,25 @@ - A language model that uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions improving systematic generalization abilities. + + + + + + + + + + + + + + + + + + + A deep neural network that uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions improving systematic generalization abilities. RLM Compositional generalization @@ -5981,8 +6372,8 @@ - A language model that uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions improving systematic generalization abilities. - https://en.wikipedia.org/wiki/Recurrent_neural_network + A deep neural network that uses recursive neural network architectures like TreeLSTMs to learn syntactic composition functions improving systematic generalization abilities. + @@ -6001,7 +6392,7 @@ A deep neural network that recursively applies weights over structured input to generate structured or scalar predictions. - https://en.wikipedia.org/wiki/Recursive_neural_network + @@ -6009,7 +6400,6 @@ - A set of statistical processes for estimating the relationships between a dependent variable and one or more independent variables. Regression analysis @@ -6021,7 +6411,7 @@ A set of statistical processes for estimating the relationships between a dependent variable and one or more independent variables. - https://en.wikipedia.org/wiki/Regression_analysis + @@ -6032,13 +6422,14 @@ A layer that applies penalties on layer parameters or layer activity during optimization summed into the loss function that the network optimizes. + Regularizers allow you to apply penalties on layer parameters or layer activity during optimization. These penalties are summed into the loss function that the network optimizes. Regularization penalties are applied on a per-layer basis. Regularization Layer A layer that applies penalties on layer parameters or layer activity during optimization summed into the loss function that the network optimizes. - https://keras.io/api/layers/regularizers/ + @@ -6046,16 +6437,16 @@ - - A type of machine learning focused on methods that do not require labeled input/output pairs or explicit correction of sub-optimal actions focusing instead on balancing exploration and exploitation to optimize performance over time. + + A machine learning task focused on methods that do not require labeled input/output pairs or explicit correction of sub-optimal actions focusing instead on balancing exploration and exploitation to optimize performance over time. Reinforcement Learning - A type of machine learning focused on methods that do not require labeled input/output pairs or explicit correction of sub-optimal actions focusing instead on balancing exploration and exploitation to optimize performance over time. - https://en.wikipedia.org/wiki/Reinforcement_learning + A machine learning task focused on methods that do not require labeled input/output pairs or explicit correction of sub-optimal actions focusing instead on balancing exploration and exploitation to optimize performance over time. + @@ -6070,6 +6461,7 @@ decision transformers reward modeling + An RL-LLM is a language model fine-tuned using reinforcement learning, where the model receives rewards for generating text that satisfies certain desired properties or objectives. This can improve the quality, safety, or alignment of generated text. Reinforcement Learning LLM @@ -6079,15 +6471,16 @@ - A layer that repeats the input n times. + A reshaping layer that repeats the input n times. + Repeats the input n times. RepeatVector Layer - A layer that repeats the input n times. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/RepeatVector + A reshaping layer that repeats the input n times. + @@ -6098,14 +6491,14 @@ A selection and sampling bias due to non-random sampling of subgroups making trends non-generalizable to new populations. + Bias due to non-random sampling of subgroups, making trends non-generalizable to new populations. Representation Bias A selection and sampling bias due to non-random sampling of subgroups making trends non-generalizable to new populations. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -6117,13 +6510,14 @@ A deep neural network that discovers representations required for feature detection or classification from raw data. Feature Learning + Discovering representations required for feature detection or classification from raw data. Representation Learning A deep neural network that discovers representations required for feature detection or classification from raw data. - https://en.wikipedia.org/wiki/Feature_learning + @@ -6132,15 +6526,15 @@ - A preprocessing layer that rescales input values to a new range. + An image preprocessing layer that rescales input values to a new range. Rescaling Layer - A preprocessing layer that rescales input values to a new range. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Rescaling + An image preprocessing layer that rescales input values to a new range. + @@ -6149,15 +6543,15 @@ - A layer that reshapes the inputs into the given shape. + A reshaping layer that reshapes the inputs into the given shape. Reshape Layer - A layer that reshapes the inputs into the given shape. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Reshape + A reshaping layer that reshapes the inputs into the given shape. + @@ -6169,13 +6563,14 @@ A layer that is used to change the shape of the input. Reshape Layer + Reshape layers are used to change the shape of the input. Reshaping Layer A layer that is used to change the shape of the input. - https://keras.io/api/layers/reshaping_layers/reshape/ + @@ -6184,6 +6579,36 @@ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + A deep neural network that employs skip connections to bypass layers facilitating learning of residual functions. DRN Deep Residual Network @@ -6197,7 +6622,7 @@ A deep neural network that employs skip connections to bypass layers facilitating learning of residual functions. - https://en.wikipedia.org/wiki/Residual_neural_network + @@ -6206,15 +6631,16 @@ - A preprocessing layer that resizes images to a target size. + An image preprocessing layer that resizes images to a target size. + A preprocessing layer which resizes images. This layer resizes an image input to a target height and width. The input should be a 4D (batched) or 3D (unbatched) tensor in "channels_last" format. Input pixel values can be of any range (e.g. [0., 1.) or [0, 255]) and of interger or floating point dtype. By default, the layer will output floats. This layer can be called on tf.RaggedTensor batches of input images of distinct sizes, and will resize the outputs to dense tensors of uniform size. Resizing Layer - A preprocessing layer that resizes images to a target size. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Resizing + An image preprocessing layer that resizes images to a target size. + @@ -6223,6 +6649,18 @@ + + + + + + + + + + + + A Boltzmann machine network that learns the probability distribution of its input data. RBM @@ -6233,7 +6671,7 @@ A Boltzmann machine network that learns the probability distribution of its input data. - https://en.wikipedia.org/wiki/Restricted_Boltzmann_machine + @@ -6242,7 +6680,7 @@ - A LLM which combines a pre-trained language model with a retrieval system that can access external knowledge sources. This allows the model to condition its generation on relevant retrieved knowledge, improving factual accuracy and knowledge grounding. + A large language model which combines a pre-trained language model with a retrieval system that can access external knowledge sources. This allows the model to condition its generation on relevant retrieved knowledge, improving factual accuracy and knowledge grounding. Retrieval-Augmented Large Language Model knowledge grounding open-book question answering @@ -6256,15 +6694,15 @@ - A regression analysis method that estimates the coefficients of multiple regression models in scenarios where the independent variables are highly correlated. + A regression analysis that estimates the coefficients of multiple regression models in scenarios where the independent variables are highly correlated. Ridge Regression - A regression analysis method that estimates the coefficients of multiple regression models in scenarios where the independent variables are highly correlated. - https://en.wikipedia.org/wiki/Ridge_regression + A regression analysis that estimates the coefficients of multiple regression models in scenarios where the independent variables are highly correlated. + @@ -6272,18 +6710,19 @@ - - An activation function that multiplies scale (> 1) with the output of the ELU function to ensure a slope larger than one for positive inputs. + + A mathematical function that multiplies scale (> 1) with the output of the ELU function to ensure a slope larger than one for positive inputs. SELU Scaled Exponential Linear Unit + SELU Function - An activation function that multiplies scale (> 1) with the output of the ELU function to ensure a slope larger than one for positive inputs. - https://www.tensorflow.org/api_docs/python/tf/keras/activations/selu + A mathematical function that multiplies scale (> 1) with the output of the ELU function to ensure a slope larger than one for positive inputs. + @@ -6309,14 +6748,14 @@ Selection Bias Selection Effect + Bias introduced by non-random selection of individuals, groups, or data, failing to ensure representativeness. Selection And Sampling Bias A computational bias introduced by non-random selection of individuals groups or data failing to ensure representativeness. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -6333,8 +6772,7 @@ An individual bias characterized by the tendency to selectively adopt algorithmic advice that matches pre-existing beliefs and stereotypes. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -6343,8 +6781,7 @@ - A LLM which learns rich representations by solving pretext tasks that involve predicting parts of the input from other observed parts of the data, without relying on human-annotated labels. - Self-Supervised LLM + A large language model which learns rich representations by solving pretext tasks that involve predicting parts of the input from other observed parts of the data, without relying on human-annotated labels. Pretext tasks Self-Supervised LLM @@ -6355,16 +6792,16 @@ - - A machine learning that is intermediate between supervised and unsupervised learning and predicts parts of the input data from other observed parts without relying on human-annotated labels. + + A machine learning task that is intermediate between supervised and unsupervised learning and predicts parts of the input data from other observed parts without relying on human-annotated labels. Self-supervised Learning - A machine learning that is intermediate between supervised and unsupervised learning and predicts parts of the input data from other observed parts without relying on human-annotated labels. - https://en.wikipedia.org/wiki/Self-supervised_learning + A machine learning task that is intermediate between supervised and unsupervised learning and predicts parts of the input data from other observed parts without relying on human-annotated labels. + @@ -6373,7 +6810,7 @@ - A LLM which combines self-supervised pretraining on unlabeled data with supervised fine-tuning on labeled task data. + A large language model which combines self-supervised pretraining on unlabeled data with supervised fine-tuning on labeled task data. Semi-Supervised Large Language Model self-training @@ -6386,16 +6823,17 @@ - A layer that performs depthwise separable 1D convolution. + A convolutional layer that performs depthwise separable 1D convolution. SeparableConv1D Layer + Depthwise separable 1D convolution. This layer performs a depthwise convolution that acts separately on channels, followed by a pointwise convolution that mixes channels. If use_bias is True and a bias initializer is provided, it adds a bias vector to the output. It then optionally applies an activation function to produce the final output.a SeparableConvolution1D Layer - A layer that performs depthwise separable 1D convolution. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/SeparableConv1D + A convolutional layer that performs depthwise separable 1D convolution. + @@ -6404,16 +6842,17 @@ - A layer that performs depthwise separable 2D convolution. + A convolutional layer that performs depthwise separable 2D convolution. SeparableConv2D Layer + Depthwise separable 2D convolution. Separable convolutions consist of first performing a depthwise spatial convolution (which acts on each input channel separately) followed by a pointwise convolution which mixes the resulting output channels. The depth_multiplier argument controls how many output channels are generated per input channel in the depthwise step. Intuitively, separable convolutions can be understood as a way to factorize a convolution kernel into two smaller kernels, or as an extreme version of an Inception block. SeparableConvolution2D Layer - A layer that performs depthwise separable 2D convolution. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/SeparableConv2D + A convolutional layer that performs depthwise separable 2D convolution. + @@ -6421,16 +6860,19 @@ - - An activation function that applies the sigmoid activation function sigmoid(x) = 1 / (1 + exp(-x)) always returning a value between 0 and 1. + + A mathematical function that applies the sigmoid activation function sigmoid(x) = 1 / (1 + exp(-x)) always returning a value between 0 and 1. + tore + + Applies the sigmoid activation function sigmoid(x) = 1 / (1 + exp(-x)). For small values (<-5), sigmoid returns a value close to zero, and for large values (>5) the result of the function gets close to 1. Sigmoid is equivalent to a 2-element Softmax, where the second element is assumed to be zero. The sigmoid function always returns a value between 0 and 1. Sigmoid Function - An activation function that applies the sigmoid activation function sigmoid(x) = 1 / (1 + exp(-x)) always returning a value between 0 and 1. - https://www.tensorflow.org/api_docs/python/tf/keras/activations/sigmoid + A mathematical function that applies the sigmoid activation function sigmoid(x) = 1 / (1 + exp(-x)) always returning a value between 0 and 1. + @@ -6441,13 +6883,14 @@ A layer that processes one step within the whole time sequence input for a SimpleRNN layer. + Cell class for SimpleRNN. This class processes one step within the whole time sequence input, whereas tf.keras.layer.SimpleRNN processes the whole sequence. SimpleRNNCell Layer A layer that processes one step within the whole time sequence input for a SimpleRNN layer. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/SimpleRNNCell + @@ -6458,13 +6901,14 @@ A recurrent layer that implements a fully-connected RNN where the output is to be fed back to input. + Fully-connected RNN where the output is to be fed back to input. SimpleRNN Layer A recurrent layer that implements a fully-connected RNN where the output is to be fed back to input. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/SimpleRNN + @@ -6473,7 +6917,7 @@ - #N/A + A selection and sampling bias where the association between two variables changes when controlling for another variable. Simpson's Paradox Simpon's Paradox Bias @@ -6481,9 +6925,8 @@ - #N/A - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + A selection and sampling bias where the association between two variables changes when controlling for another variable. + @@ -6491,18 +6934,16 @@ - - A bias characterized by being for or against groups or individuals based on social identities demographic factors or immutable physical characteristics often manifesting as stereotypes. - Social Bias + + A systemic bias characterized by being for or against groups or individuals based on social identities demographic factors or immutable physical characteristics often manifesting as stereotypes. Societal Bias - A bias characterized by being for or against groups or individuals based on social identities demographic factors or immutable physical characteristics often manifesting as stereotypes. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + A systemic bias characterized by being for or against groups or individuals based on social identities demographic factors or immutable physical characteristics often manifesting as stereotypes. + @@ -6510,16 +6951,18 @@ - - An activation function where the elements of the output vector are in range (0 1) and sum to 1 and each vector is handled independently. + + A mathematical function where the elements of the output vector are in range (0 1) and sum to 1 and each vector is handled independently. + + The elements of the output vector are in range (0, 1) and sum to 1. Each vector is handled independently. The axis argument sets which axis of the input the function is applied along. Softmax is often used as the activation for the last layer of a classification network because the result could be interpreted as a probability distribution. The softmax of each vector x is computed as exp(x) / tf.reduce_sum(exp(x)). The input values in are the log-odds of the resulting probability. Softmax Function - An activation function where the elements of the output vector are in range (0 1) and sum to 1 and each vector is handled independently. - https://www.tensorflow.org/api_docs/python/tf/keras/activations/softmax + A mathematical function where the elements of the output vector are in range (0 1) and sum to 1 and each vector is handled independently. + @@ -6530,13 +6973,14 @@ An activation layer that applies the softmax function to the inputs. + Softmax activation function. Softmax Layer An activation layer that applies the softmax function to the inputs. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Softmax + @@ -6544,16 +6988,18 @@ - - An activation function that is softplus(x) = log(exp(x) + 1). + + A mathematical function that is softplus(x) = log(exp(x) + 1). + + softplus(x) = log(exp(x) + 1) Softplus Function - An activation function that is softplus(x) = log(exp(x) + 1). - https://www.tensorflow.org/api_docs/python/tf/keras/activations/softplus + A mathematical function that is softplus(x) = log(exp(x) + 1). + @@ -6561,16 +7007,18 @@ - - An activation function that is softsign(x) = x / (abs(x) + 1). + + A mathematical function that is softsign(x) = x / (abs(x) + 1). + + softsign(x) = x / (abs(x) + 1) Softsign Function - An activation function that is softsign(x) = x / (abs(x) + 1). - https://www.tensorflow.org/api_docs/python/tf/keras/activations/softsign + A mathematical function that is softsign(x) = x / (abs(x) + 1). + @@ -6579,6 +7027,24 @@ + + + + + + + + + + + + + + + + + + An autoencoder network with more hidden units than inputs that constrains only a few hidden units to be active at once. SAE Sparse AE @@ -6618,7 +7084,7 @@ A representation learning network that finds sparse representations of input data as a linear combination of basic elements and identifies those elements. - https://en.wikipedia.org/wiki/Sparse_dictionary_learning + @@ -6629,13 +7095,14 @@ A regularization layer that performs the same function as Dropout but drops entire 1D feature maps instead of individual elements. + Spatial 1D version of Dropout. This version performs the same function as Dropout, however, it drops entire 1D feature maps instead of individual elements. If adjacent frames within feature maps are strongly correlated (as is normally the case in early convolution layers) then regular dropout will not regularize the activations and will otherwise just result in an effective Learning rate decrease. In this case, SpatialDropout1D will help promote independence between feature maps and should be used instead. SpatialDropout1D Layer A regularization layer that performs the same function as Dropout but drops entire 1D feature maps instead of individual elements. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/SpatialDropout1D + @@ -6646,13 +7113,14 @@ A regularization layer that performs the same function as Dropout but drops entire 2D feature maps instead of individual elements. + Spatial 2D version of Dropout. This version performs the same function as Dropout, however, it drops entire 2D feature maps instead of individual elements. If adjacent pixels within feature maps are strongly correlated (as is normally the case in early convolution layers) then regular dropout will not regularize the activations and will otherwise just result in an effective Learning rate decrease. In this case, SpatialDropout2D will help promote independence between feature maps and should be used instead.a SpatialDropout2D Layer A regularization layer that performs the same function as Dropout but drops entire 2D feature maps instead of individual elements. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/SpatialDropout2D + @@ -6663,13 +7131,14 @@ A regularization layer that performs the same function as Dropout but drops entire 3D feature maps instead of individual elements. + Spatial 3D version of Dropout. This version performs the same function as Dropout, however, it drops entire 3D feature maps instead of individual elements. If adjacent voxels within feature maps are strongly correlated (as is normally the case in early convolution layers) then regular dropout will not regularize the activations and will otherwise just result in an effective Learning rate decrease. In this case, SpatialDropout3D will help promote independence between feature maps and should be used instead. SpatialDropout3D Layer A regularization layer that performs the same function as Dropout but drops entire 3D feature maps instead of individual elements. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/SpatialDropout3D + @@ -6678,15 +7147,15 @@ - A regression analysis method used to model spatial relationships. + A regression analysis used to model spatial relationships. Spatial Regression - A regression analysis method used to model spatial relationships. - https://gisgeography.com/spatial-regression-models-arcgis/ + A regression analysis used to model spatial relationships. + @@ -6695,16 +7164,7 @@ - A hidden layer that makes connections to an additional, heterogeneous hidden layer; modeled after biological neural networks. - - Spiking Hidden Layer - - - - A hidden layer that makes connections to an additional, heterogeneous hidden layer; modeled after biological neural networks. - https://doi.org/10.1016/S0893-6080(97)00011-7 - @@ -6714,13 +7174,14 @@ A layer that allows a stack of RNN cells to behave as a single cell. + Wrapper allowing a stack of RNN cells to behave as a single cell. Used to implement efficient stacked RNNs. StackedRNNCells Layer A layer that allows a stack of RNN cells to behave as a single cell. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/StackedRNNCells + @@ -6738,8 +7199,7 @@ An individual bias where people search only where it is easiest to look. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -6756,7 +7216,7 @@ A categorical features preprocessing layer that maps string features to integer indices. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/StringLookup + @@ -6767,13 +7227,14 @@ A merging layer that subtracts two inputs. + Layer that subtracts two inputs. It takes as input a list of tensors of size 2, both of the same shape, and returns a single tensor, (inputs[0] - inputs[1]), also of the same shape. Subtract Layer A merging layer that subtracts two inputs. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Subtract + @@ -6782,7 +7243,7 @@ - The process of dividing text into subword units which are smaller than words but larger than individual characters to improve the efficiency and effectiveness of natural language processing models by capturing meaningful subunits of words. + A data preparation that divides text into subword units which are smaller than words but larger than individual characters to improve the efficiency and effectiveness of natural language processing models by capturing meaningful subunits of words. Fragmentation Part-word Division Byte Pair Encoding @@ -6800,14 +7261,14 @@ A bias characterized by the tendency to continue an endeavor due to previously invested resources despite costs outweighing benefits. Sunk Cost Fallacy + The tendency to continue an endeavor due to previously invested resources, despite costs outweighing benefits. Sunk Cost Fallacy Bias A bias characterized by the tendency to continue an endeavor due to previously invested resources despite costs outweighing benefits. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -6829,7 +7290,7 @@ A biclustering task focused on methods that simultaneously cluster the rows and columns of a labeled matrix considering data labels to enhance cluster coherence. - https://en.wikipedia.org/wiki/Biclustering + @@ -6838,16 +7299,16 @@ - A clustering task focused on methods that group labeled objects such that objects in the same group have similar labels relative to those in other groups. - Cluster analysis + A clustering focused on methods that group labeled objects such that objects in the same group have similar labels relative to those in other groups. + Supervised Cluster Analysis Supervised Clustering - A clustering task focused on methods that group labeled objects such that objects in the same group have similar labels relative to those in other groups. - https://en.wikipedia.org/wiki/Cluster_analysis + A clustering focused on methods that group labeled objects such that objects in the same group have similar labels relative to those in other groups. + @@ -6855,16 +7316,16 @@ - - A type of machine learning focused on methods that learn a function mapping input to output based on example input-output pairs. + + A machine learning task focused on methods that learn a function mapping input to output based on example input-output pairs. Supervised Learning - A type of machine learning focused on methods that learn a function mapping input to output based on example input-output pairs. - https://en.wikipedia.org/wiki/Supervised_learning + A machine learning task focused on methods that learn a function mapping input to output based on example input-output pairs. + @@ -6873,6 +7334,24 @@ + + + + + + + + + + + + + + + + + + A network with supervised learning models for classification and regression that maps training examples to points in space maximizing the gap between categories. SVM SVN @@ -6885,7 +7364,7 @@ A network with supervised learning models for classification and regression that maps training examples to points in space maximizing the gap between categories. - https://en.wikipedia.org/wiki/Support-vector_machine + @@ -6893,7 +7372,7 @@ - + A machine learning task focused on methods for analyzing the expected duration of time until one or more events occur such as death in biological organisms or failure in mechanical systems. Survival Analysis @@ -6902,7 +7381,7 @@ A machine learning task focused on methods for analyzing the expected duration of time until one or more events occur such as death in biological organisms or failure in mechanical systems. - https://en.wikipedia.org/wiki/Survival_analysis + @@ -6913,14 +7392,14 @@ A processing bias characterized by the tendency to focus on items observations or people that "survive" a selection process overlooking those that did not. + The tendency to focus on items, observations, or people that "survive" a selection process, overlooking those that did not. Survivorship Bias A processing bias characterized by the tendency to focus on items observations or people that "survive" a selection process overlooking those that did not. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -6928,16 +7407,18 @@ - - An activation function that is x*sigmoid(x) a smooth non-monotonic function that consistently matches or outperforms ReLU on deep networks. + + A mathematical function that is x*sigmoid(x) a smooth non-monotonic function that consistently matches or outperforms ReLU on deep networks. + + x*sigmoid(x). It is a smooth, non-monotonic function that consistently matches or outperforms ReLU on deep networks, it is unbounded above and bounded below. Swish Function - An activation function that is x*sigmoid(x) a smooth non-monotonic function that consistently matches or outperforms ReLU on deep networks. - https://www.tensorflow.org/api_docs/python/tf/keras/activations/swish + A mathematical function that is x*sigmoid(x) a smooth non-monotonic function that consistently matches or outperforms ReLU on deep networks. + @@ -6949,13 +7430,14 @@ A network that is a type of recurrent neural network where connections between units are symmetrical with equal weights in both directions. SCN + Symmetrically connected networks are a type of recurrent neural network where connections between units are symmetrical, meaning they have equal weights in both directions. This structure allows the network to maintain consistent information flow and equilibrium. Symmetrically Connected Network A network that is a type of recurrent neural network where connections between units are symmetrical with equal weights in both directions. - https://ieeexplore.ieee.org/document/287176 + @@ -6967,13 +7449,14 @@ A batch normalization layer that applies synchronous Batch Normalization across multiple devices. SyncBatchNorm + Applies Batch Normalization over a N-Dimensional input (a mini-batch of [N-2]D inputs with additional channel dimension) as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift . SyncBatchNorm Layer A batch normalization layer that applies synchronous Batch Normalization across multiple devices. - https://pytorch.org/docs/stable/nn.html#normalization-layers + @@ -6982,18 +7465,15 @@ - A bias resulting from procedures and practices of institutions that operate in ways which result in certain social groups being advantaged or favored and others being disadvantaged or devalued. - Institutional Bias - Societal Bias + A bias resulting from procedures and practices of systems that operate in ways which result in certain social groups being advantaged or favored and others being disadvantaged or devalued. Systemic Bias - A bias resulting from procedures and practices of institutions that operate in ways which result in certain social groups being advantaged or favored and others being disadvantaged or devalued. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + A bias resulting from procedures and practices of systems that operate in ways which result in certain social groups being advantaged or favored and others being disadvantaged or devalued. + @@ -7001,17 +7481,19 @@ - - An activation function that is the hyperbolic tangent activation function. + + A mathematical function that is the hyperbolic tangent activation function. hyperbolic tangent + + Hyperbolic tangent activation function. Tanh Function - An activation function that is the hyperbolic tangent activation function. - https://www.tensorflow.org/api_docs/python/tf/keras/activations/tanh + A mathematical function that is the hyperbolic tangent activation function. + @@ -7028,8 +7510,7 @@ A selection and sampling bias arising from differences in populations and behaviors over time. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -7046,7 +7527,7 @@ A layer that performs text data preprocessing operations. - https://keras.io/guides/preprocessing_layers/ + @@ -7055,15 +7536,15 @@ - A preprocessing layer that maps text features to integer sequences. + A text preprocessing layer that maps text features to integer sequences. TextVectorization Layer - A preprocessing layer that maps text features to integer sequences. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/TextVectorization + A text preprocessing layer that maps text features to integer sequences. + @@ -7077,6 +7558,12 @@ Threshold Autoregressive + + + + A model that allows for different autoregressive processes depending on the regime or state of the time series, enabling the capture of nonlinear behaviors. + + @@ -7086,13 +7573,14 @@ An activation layer that applies the thresholded rectified linear unit function element-wise. + Thresholded Rectified Linear Unit. ThresholdedReLU Layer An activation layer that applies the thresholded rectified linear unit function element-wise. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/ThresholdedReLU + @@ -7101,15 +7589,16 @@ - A wrapper layer that applies a layer to every temporal slice of an input. + A recurrent layer that applies a layer to every temporal slice of an input. + This wrapper allows to apply a layer to every temporal slice of an input. Every input should be at least 3D, and the dimension of index one of the first input will be considered to be the temporal dimension. Consider a batch of 32 video samples, where each sample is a 128x128 RGB image with channels_last data format, across 10 timesteps. The batch input shape is (32, 10, 128, 128, 3). You can then use TimeDistributed to apply the same Conv2D layer to each of the 10 timesteps, independently: TimeDistributed Layer - A wrapper layer that applies a layer to every temporal slice of an input. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/TimeDistributed + A recurrent layer that applies a layer to every temporal slice of an input. + @@ -7117,16 +7606,17 @@ - + A machine learning task focused on methods for analyzing time series data to extract meaningful statistics and characteristics. + Methods for analyzing time series data to extract meaningful statistics and characteristics. Time Series Analysis A machine learning task focused on methods for analyzing time series data to extract meaningful statistics and characteristics. - https://en.wikipedia.org/wiki/Time_series + @@ -7134,16 +7624,17 @@ - + A machine learning task focused on methods that predict future values based on previously observed values. + Methods that predict future values based on previously observed values. Time Series Forecasting A machine learning task focused on methods that predict future values based on previously observed values. - https://en.wikipedia.org/wiki/Time_series + @@ -7152,7 +7643,7 @@ - The process of converting a sequence of text into smaller meaningful units called tokens typically words or subwords for the purpose of analysis or processing by language models. + A data preparation that converts a sequence of text into smaller meaningful units called tokens typically words or subwords for the purpose of analysis or processing by language models. Lexical Analysis Text Segmentation @@ -7165,7 +7656,7 @@ - The methodologies and approaches used to train machine learning models including techniques such as supervised learning unsupervised learning reinforcement learning and transfer learning aimed at optimizing model performance. + A preprocessing used to train machine learning models including techniques such as supervised learning unsupervised learning reinforcement learning and transfer learning aimed at optimizing model performance. Instructional Methods Learning Techniques @@ -7177,16 +7668,16 @@ - - A type of machine learning focused on methods that reuse or transfer information from previously learned tasks to facilitate the learning of new tasks. + + A machine learning task focused on methods that reuse or transfer information from previously learned tasks to facilitate the learning of new tasks. Transfer Learning - A type of machine learning focused on methods that reuse or transfer information from previously learned tasks to facilitate the learning of new tasks. - https://en.wikipedia.org/wiki/Transfer_learning + A machine learning task focused on methods that reuse or transfer information from previously learned tasks to facilitate the learning of new tasks. + @@ -7217,7 +7708,7 @@ A transformer language model with large training corpuses and sets of parameters that uses the transformer architecture based on multi-head attention mechanisms allowing it to contextualize tokens within a context window for effective language understanding and generation. - https://en.wikipedia.org/wiki/Transformer_(deep_learning_architecture) + @@ -7235,8 +7726,7 @@ A language model that uses the transformer architecture based on multi-head attention mechanisms allowing it to contextualize tokens within a context window for effective language understanding and generation. - - https://arxiv.org/abs/1706.03762 + @@ -7247,13 +7737,14 @@ A deep neural network that utilizes attention mechanisms to weigh the significance of input data. + A transformer network utilizes attention mechanisms to weigh the significance of each part of the input data, widely used in natural language processing (NLP) and computer vision (CV). Transformer Network A deep neural network that utilizes attention mechanisms to weigh the significance of input data. - https://en.wikipedia.org/wiki/Transformer_(machine_Learning_model) + @@ -7264,14 +7755,14 @@ A selection and sampling bias favoring groups better represented in training data due to less prediction uncertainty. + Bias favoring groups better represented in training data, due to less prediction uncertainty. Uncertainty Bias A selection and sampling bias favoring groups better represented in training data due to less prediction uncertainty. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -7280,15 +7771,16 @@ - A normalization layer that normalizes a batch of inputs so that each input in the batch has a L2 norm equal to 1. + A recurrent layer that normalizes a batch of inputs so that each input in the batch has a L2 norm equal to 1. + Unit normalization layer. Normalize a batch of inputs so that each input in the batch has a L2 norm equal to 1 (across the axes specified in axis). UnitNormalization Layer - A normalization layer that normalizes a batch of inputs so that each input in the batch has a L2 norm equal to 1. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/UnitNormalization + A recurrent layer that normalizes a batch of inputs so that each input in the batch has a L2 norm equal to 1. + @@ -7298,11 +7790,11 @@ A biclustering task focused on methods that simultaneously cluster the rows and columns of an unlabeled input matrix to identify submatrices with coherent patterns. - Block Clustering - Co-clustering - Joint Clustering - Two-mode Clustering - Two-way Clustering + Unsupervised Block Clustering + Unsupervised Co-clustering + Unsupervised Joint Clustering + Unsupervised Two-mode Clustering + Unsupervised Two-way Clustering Unsupervised Biclustering @@ -7310,7 +7802,7 @@ A biclustering task focused on methods that simultaneously cluster the rows and columns of an unlabeled input matrix to identify submatrices with coherent patterns. - https://en.wikipedia.org/wiki/Biclustering + @@ -7319,16 +7811,16 @@ - A clustering task focused on methods that group a set of unlabeled objects such that objects in the same group are more similar to each other than to those in other groups. - Cluster analysis + A clustering focused on methods that group a set of unlabeled objects such that objects in the same group are more similar to each other than to those in other groups. + Unsupervised Cluster Analysis Unsupervised Clustering - A clustering task focused on methods that group a set of unlabeled objects such that objects in the same group are more similar to each other than to those in other groups. - https://en.wikipedia.org/wiki/Cluster_analysis + A clustering focused on methods that group a set of unlabeled objects such that objects in the same group are more similar to each other than to those in other groups. + @@ -7349,16 +7841,17 @@ - - A type of machine learning focused on algorithms that learn patterns from unlabeled data. + + A machine learning task focused on algorithms that learn patterns from unlabeled data. + Algorithms that learn patterns from unlabeled data. Unsupervised Learning - A type of machine learning focused on algorithms that learn patterns from unlabeled data. - https://en.wikipedia.org/wiki/Unsupervised_learning + A machine learning task focused on algorithms that learn patterns from unlabeled data. + @@ -7370,13 +7863,14 @@ A network that initializes a discriminative neural net from one trained using an unsupervised criterion. UPN + Unsupervised pre-training initializes a discriminative neural net from one trained using an unsupervised criterion, aiding in optimization and overfitting issues. Unsupervised Pretrained Network A network that initializes a discriminative neural net from one trained using an unsupervised criterion. - https://metacademy.org/graphs/concepts/unsupervised_pre_training#:~:text=Unsupervised%20pre%2Dtraining%20initializes%20a,optimization%20and%20the%20overfitting%20issues + @@ -7385,15 +7879,16 @@ - A layer that upsamples the input by repeating each temporal step size times along the time axis. + A reshaping layer that upsamples the input by repeating each temporal step size times along the time axis. + Upsampling layer for 1D inputs. Repeats each temporal step size times along the time axis. UpSampling1D Layer - A layer that upsamples the input by repeating each temporal step size times along the time axis. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/UpSampling1D + A reshaping layer that upsamples the input by repeating each temporal step size times along the time axis. + @@ -7404,13 +7899,14 @@ A layer that upsamples the input by repeating each row and column size times. + Upsampling layer for 2D inputs. Repeats the rows and columns of the data by size[0] and size[1] respectively. UpSampling2D Layer A layer that upsamples the input by repeating each row and column size times. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/UpSampling2D + @@ -7421,13 +7917,14 @@ A layer that upsamples the input by repeating each depth + Upsampling layer for 3D inputs. UpSampling3D Layer A layer that upsamples the input by repeating each depth - https://www.tensorflow.org/api_docs/python/tf/keras/layers/UpSampling3D + @@ -7439,14 +7936,14 @@ A computational bias characterized by inappropriately analyzing ambiguous stimuli scenarios and events. Interpretive Bias + Bias inappropriately analyzing ambiguous stimuli, scenarios, and events. Use And Interpretation Bias A computational bias characterized by inappropriately analyzing ambiguous stimuli scenarios and events. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -7457,14 +7954,14 @@ An individual bias arising when a user imposes their own biases during interaction with data output results etc. + Bias arising when a user imposes their own biases during interaction with data, output, results, etc. User Interaction Bias An individual bias arising when a user imposes their own biases during interaction with data output results etc. - GTP-4o with Seppala et al. 2017 - https://doi.org/10.6028/NIST.SP.1270 + @@ -7473,6 +7970,24 @@ + + + + + + + + + + + + + + + + + + An autoencoder network that imposes a probabilistic structure on the latent space for unsupervised learning. VAE @@ -7498,7 +8013,7 @@ - The technique of limiting the number of unique tokens in a language model's vocabulary by merging or eliminating less frequent tokens thereby optimizing computational efficiency and resource usage. + A data preparation that limits the number of unique tokens in a language model's vocabulary by merging or eliminating less frequent tokens thereby optimizing computational efficiency and resource usage. Lexical Simplification Lexicon Pruning Vocabulary Condensation @@ -7512,9 +8027,6 @@ - A layer of values to be applied to other cells or neurons in a network. - - Weighted Layer @@ -7523,15 +8035,16 @@ - An abstract base class for wrappers that augment the functionality of another layer. + A layer that augment the functionality of another layer. + Abstract wrapper base class. Wrappers take another layer and augment it in various ways. Do not use this class as a layer, it is only an abstract base class. Two usable wrappers are the TimeDistributed and Bidirectional wrappers. Wrapper Layer - An abstract base class for wrappers that augment the functionality of another layer. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/Wrapper + A layer that augment the functionality of another layer. + @@ -7540,15 +8053,16 @@ - A layer that zero-pads the input along the time axis. + A reshaping layer that zero-pads the input along the time axis. + Zero-padding layer for 1D input (e.g. temporal sequence). ZeroPadding1D Layer - A layer that zero-pads the input along the time axis. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/ZeroPadding1D + A reshaping layer that zero-pads the input along the time axis. + @@ -7557,15 +8071,16 @@ - A layer that zero-pads the input along the height and width dimensions. + A reshaping layer that zero-pads the input along the height and width dimensions. + Zero-padding layer for 2D input (e.g. picture). This layer can add rows and columns of zeros at the top, bottom, left and right side of an image tensor. ZeroPadding2D Layer - A layer that zero-pads the input along the height and width dimensions. - https://www.tensorflow.org/api_docs/python/tf/keras/layers/ZeroPadding2D + A reshaping layer that zero-pads the input along the height and width dimensions. + @@ -7574,15 +8089,16 @@ - A layer that zero-pads the input along the depth + A reshaping layer that zero-pads the input along the depth + Zero-padding layer for 3D data (spatial or spatio-temporal). ZeroPadding3D Layer - A layer that zero-pads the input along the depth - https://www.tensorflow.org/api_docs/python/tf/keras/layers/ZeroPadding3D + A reshaping layer that zero-pads the input along the depth + @@ -7591,7 +8107,7 @@ - A LLM which performs tasks or understands concepts it has not explicitly been trained on, demonstrating a high degree of generalization and understanding. + A large language model which performs tasks or understands concepts it has not explicitly been trained on, demonstrating a high degree of generalization and understanding. Zero-Shot LLM zero-shot learning @@ -7613,7 +8129,7 @@ A deep neural network that predicts classes at test time from classes not observed during training. - https://en.wikipedia.org/wiki/Zero-shot_learning + @@ -7621,10 +8137,27 @@ - - A machine learning designed to learn continuous feature representations for nodes in a graph by optimizing a neighborhood-preserving objective. + + + + + + + + + + + + + + + + + + + + A machine learning task designed to learn continuous feature representations for nodes in a graph by optimizing a neighborhood-preserving objective. N2V - node2vec Layers: Input, Hidden, Output node2vec @@ -7632,8 +8165,8 @@ - A machine learning designed to learn continuous feature representations for nodes in a graph by optimizing a neighborhood-preserving objective. - https://en.wikipedia.org/wiki/Node2vec + A machine learning task designed to learn continuous feature representations for nodes in a graph by optimizing a neighborhood-preserving objective. + @@ -7641,7 +8174,25 @@ - + + + + + + + + + + + + + + + + + + + A node2vec that predicts the current node from a window of surrounding context nodes, with the order of context nodes not influencing prediction. N2V-CBOW CBOW @@ -7653,7 +8204,7 @@ A node2vec that predicts the current node from a window of surrounding context nodes, with the order of context nodes not influencing prediction. - https://en.wikipedia.org/wiki/Word2vec + @@ -7662,6 +8213,24 @@ + + + + + + + + + + + + + + + + + + A node2vec that uses the current node to predict the surrounding window of context nodes, weighing nearby context nodes more heavily than distant ones. N2V-SkipGram SkipGram @@ -7673,7 +8242,7 @@ A node2vec that uses the current node to predict the surrounding window of context nodes, weighing nearby context nodes more heavily than distant ones. - https://en.wikipedia.org/wiki/Word2vec + @@ -7692,7 +8261,7 @@ A dimensionality reduction for visualizing high-dimensional data by giving each datapoint a location in a two or three-dimensional map. - https://en.wikipedia.org/wiki/T-distributed_stochastic_neighbor_embedding + @@ -7700,10 +8269,27 @@ - - A machine learning that generates distributed representations of words by training a shallow neural network model, which aims to predict the context of each word within a corpus. This algorithm captures semantic meanings of words through their contextual usage in the text. + + + + + + + + + + + + + + + + + + + + A machine learning task that generates distributed representations of words by training a shallow neural network model, which aims to predict the context of each word within a corpus. This algorithm captures semantic meanings of words through their contextual usage in the text. W2V - word2vec Layers: Input, Hidden, Output word2vec @@ -7711,8 +8297,8 @@ - A machine learning that generates distributed representations of words by training a shallow neural network model, which aims to predict the context of each word within a corpus. This algorithm captures semantic meanings of words through their contextual usage in the text. - https://en.wikipedia.org/wiki/Word2vec + A machine learning task that generates distributed representations of words by training a shallow neural network model, which aims to predict the context of each word within a corpus. This algorithm captures semantic meanings of words through their contextual usage in the text. + @@ -7721,6 +8307,24 @@ + + + + + + + + + + + + + + + + + + A word2vec that predicts the current word from a window of surrounding context words, ignoring the order of context words. W2V-CBOW CBOW @@ -7732,7 +8336,7 @@ A word2vec that predicts the current word from a window of surrounding context words, ignoring the order of context words. - https://en.wikipedia.org/wiki/Word2vec + @@ -7741,6 +8345,24 @@ + + + + + + + + + + + + + + + + + + A word2vec that predicts surrounding context words from the current word, giving more weight to nearby context words than distant ones. W2V-SkipGram SkipGram @@ -7752,11 +8374,186 @@ A word2vec that predicts surrounding context words from the current word, giving more weight to nearby context words than distant ones. - https://en.wikipedia.org/wiki/Word2vec + + + + + + + + + A layer that adds inputs from one or more other layers to cells or neurons of a target layer. + + Addition Layer + + + An input layer that receives values from another layer. + + Backfed Input Layer + + + A normalization layer that normalizes its inputs applying a transformation that maintains the mean close to 0 and the standard deviation close to 1. + BatchNorm + + Layer that normalizes its inputs. Batch normalization applies a transformation that maintains the mean output close to 0 and the output standard deviation close to 1. Importantly, batch normalization works differently during training and during inference. During training (i.e. when using fit() or when calling the layer/model with the argument training=True), the layer normalizes its output using the mean and standard deviation of the current batch of inputs. That is to say, for each channel being normalized, the layer returns gamma * (batch - mean(batch)) / sqrt(var(batch) + epsilon) + beta, where: epsilon is small constant (configurable as part of the constructor arguments), gamma is a learned scaling factor (initialized as 1), which can be disabled by passing scale=False to the constructor. beta is a learned offset factor (initialized as 0), which can be disabled by passing center=False to the constructor. During inference (i.e. when using evaluate() or predict() or when calling the layer/model with the argument training=False (which is the default), the layer normalizes its output using a moving average of the mean and standard deviation of the batches it has seen during training. That is to say, it returns gamma * (batch - self.moving_mean) / sqrt(self.moving_var + epsilon) + beta. self.moving_mean and self.moving_var are non-trainable variables that are updated each time the layer in called in training mode, as such: moving_mean = moving_mean * momentum + mean(batch) * (1 - momentum) moving_var = moving_var * momentum + var(batch) * (1 - momentum). + BatchNormalization Layer + + + + + A normalization layer that normalizes its inputs applying a transformation that maintains the mean close to 0 and the standard deviation close to 1. + + + + A layer that contains a set of filters (or kernels) parameters of which are to be learned throughout the training. + + A convolutional layer is the main building block of a CNN. It contains a set of filters (or kernels), parameters of which are to be learned throughout the training. The size of the filters is usually smaller than the actual image. Each filter convolves with the image and creates an activation map. + Convolutional Layer + + + + + A layer that contains a set of filters (or kernels) parameters of which are to be learned throughout the training. + + + A layer located between the input and output that performs nonlinear transformations of the inputs entered into the network. + + A hidden layer is located between the input and output of the algorithm, in which the function applies weights to the inputs and directs them through an activation function as the output. In short, the hidden layers perform nonlinear transformations of the inputs entered into the network. Hidden layers vary depending on the function of the neural network, and similarly, the layers may vary depending on their associated weights. + Hidden Layer + + + + + A layer located between the input and output that performs nonlinear transformations of the inputs entered into the network. + + + + A layer composed of artificial input neurons that brings the initial data into the system for further processing by subsequent layers. + + The input layer of a neural network is composed of artificial input neurons, and brings the initial data into the system for further processing by subsequent layers of artificial neurons. The input layer is the very beginning of the workflow for the artificial neural network. + Input Layer + + + + + A layer composed of artificial input neurons that brings the initial data into the system for further processing by subsequent layers. + + + + A layer that obtains the dot product of input values or subsets of input values. + + Kernel Layer + + + An input layer with a shape corresponding to that of the output layer. + + Matched Input-Output Layer + + + A layer of cells, each with an internal state or weights. + + Memory Cell Layer + + + + + A layer of cells, each with an internal state or weights. + + + + An input layer that adds noise to each value. + + Noisy Input Layer + + + + + An input layer that adds noise to each value. + + + + A layer containing the last neurons in the network that produces given outputs for the program. + + The output layer in an artificial neural network is the last layer of neurons that produces given outputs for the program. Though they are made much like other artificial neurons in the neural network, output layer neurons may be built or observed in a different way, given that they are the last “actor” nodes on the network. + Output Layer + + + + + A layer containing the last neurons in the network that produces given outputs for the program. + + + + A layer that, after taking a set of states or values as input, predicts a probability distribution of actions to take. + + Policy Layer + + + A layer that serves to mitigate the sensitivity of convolutional layers to location and spatially downsample representations. + + Pooling layers serve the dual purposes of mitigating the sensitivity of convolutional layers to location and of spatially downsampling representations. + Pooling Layer + + + + + A layer that serves to mitigate the sensitivity of convolutional layers to location and spatially downsample representations. + + + + A hidden layer that estimates the probability of a sample being within a certain category. + + Probabilistic Hidden Layer + + + An activation layer that applies the rectified linear unit function element-wise. + + Rectified Linear Unit activation function. With default values, it returns element-wise max(x, 0). + ReLU Layer + + + + + An activation layer that applies the rectified linear unit function element-wise. + + + + A layer composed of recurrent units with the number equal to the hidden size of the layer. + + Recurrent Layer + + + + + A layer composed of recurrent units with the number equal to the hidden size of the layer. + + + + A hidden layer that makes connections to an additional, heterogeneous hidden layer; modeled after biological neural networks. + + Spiking Hidden Layer + + + + + A hidden layer that makes connections to an additional, heterogeneous hidden layer; modeled after biological neural networks. + + + + A layer of values to be applied to other cells or neurons in a network. + + Weighted Layer + - + diff --git a/src/scripts/transpose_filter_statistics.py b/src/scripts/transpose_filter_statistics.py new file mode 100644 index 0000000..00134e7 --- /dev/null +++ b/src/scripts/transpose_filter_statistics.py @@ -0,0 +1,7 @@ +import sys + +prefixes = ('id', 'class_count', 'subclass_of_axiom_count', 'distinct_synonym_count') +lines = [line.strip().split('\t') for line in sys.stdin] +transposed = zip(*lines) +filtered = [row for row in transposed if row[0] in prefixes] +print('\n'.join('\t'.join(item for item in row) for row in filtered)) diff --git a/src/sparql/subclass-label-compatibility.rq b/src/sparql/subclass-label-compatibility.rq new file mode 100644 index 0000000..f4122e0 --- /dev/null +++ b/src/sparql/subclass-label-compatibility.rq @@ -0,0 +1,15 @@ +PREFIX obo: +PREFIX rdfs: + +SELECT ?parent ?parent_label ?child ?child_def +WHERE + { ?child rdfs:subClassOf ?parent + MINUS + { ?child rdfs:subClassOf } + FILTER isURI(?parent) + OPTIONAL + { ?child obo:IAO_0000115 ?child_def } + OPTIONAL + { ?parent rdfs:label ?parent_label } + } +ORDER BY ?parent ?child diff --git a/synonym-counts.rq b/synonym-counts.rq deleted file mode 100644 index b2c2209..0000000 --- a/synonym-counts.rq +++ /dev/null @@ -1,18 +0,0 @@ -prefix rdfs: -prefix owl: -prefix oio: - -select ?branch (count(distinct ?synonym) as ?synonym_count) - where { - values ?syn_type { - oio:hasExactSynonym - oio:hasRelatedSynonym - } - ?branch rdfs:subClassOf owl:Thing . - ?child_class rdfs:subClassOf* ?branch . - ?child_class ?syn_type ?synonym . -} -group by ?branch -order by ?branch - -