a predictive learning rule in single neurons

summary

this is a repository for the paper:
"Sequence anticipation and spike-timing-dependent plasticity emerge from a predictive learning rule"
M Saponati, M Vinck
Nature Communications 14, 4985 (2023).
https://doi.org/10.1038/s41467-023-40651-w

---

---

installation/dependencies

The current version of the scripts has been tested with Python 3.8. All the dependencies are listed in the environment.yml file. The project has a pip-installable package. How to set it up:

• git clone the repository
• pip install -e .

structure

this repo is structured as follows:

• ./figures/: contains the code necessary to reproduce all the figures in the paper

• ./models/ contains the Python Class of the different models

• ./scripts/ contains scripts to run the model on different types of inputs and network implementations

• ./utils/ contains the Python modules for training and the helper functions for the analysis

• environment.yml configuration file with all the dependencies listed

• setup.py python script for installation with pip

---

citation and credits

Saponati, M., Vinck, M. (2023). Sequence anticipation and spike-timing-dependent plasticity emerge from a predictive learning rule. Nature Communications, 14(1), 1-13.

@article{saponati2023sequence,
  title={Sequence anticipation and spike-timing-dependent plasticity emerge from a predictive learning rule},
  author={Saponati, Matteo and Vinck, Martin},
  journal={Nature communications},
  volume={14},
  number={1},
  pages={1--13},
  year={2023},
  publisher={Nature Publishing Group}
}