Note
To solve the questions in the Weeks (If you are interested 😊), check out the Questions folder in the Weeks. Check out the Course Info file for more details!
Note: The books that have copyrights aren't being uploaded here. The books in the Resources/Books/ directory are downloaded from the sources listed in the 'Books and Resources' section.
If you want to learn from this repository(❤️), follow the README files to stay organized and on track.
This work is licensed under the Creative Commons Attribution-ShareAlike 4.0 International License.
Refer this for more info!
-
Grokking RL: I didn't find an official resource 😢...
-
Sutton and Barto: This book can be downloaded from:
-
Thompson Sampling: This resource can be obtained from:
.
├── Resources/
│ └── Books/
├── Week1/
│ ├── Assignment/
│ ├── {Solved Files} # Files related to solved questions
│ └── README.md
├── Week2/
│ ├── Assignment/
│ ├── {Solved Files} # Files related to solved questions
│ └── README.md
.
. (Continue similarly for other weeks)
.
├── .gitignore
├── CourseInfo.md
└── README.md
This week covers the basic topics of python required for Machine Learning and Reinforcement Learning.
The basics of Reinforcement Learning are also covered in the books Grokking RL and Sutton and Barto.
This week covers the basics of Reinforcement Learning, mainly covered from Grokking RL.
It also has an assignment to cover the basics of writing a game in python using the pygame module.
This week covers the algorithms to deal with Prediction and Control problems.
This week covers the implementation of the Multi-armed bandit problem implemented in python.
This week covers Temporal Difference Learning, Q Learning and Eligibility Traces along with a simple game and using Q-learning to "solve" the environment.
We have reached the ending! We will be implementing this paper to create a Chess Engine based on Deep RL.
Tip
There were some dead weeks also in between in my SoC. Considering those also would make this an ~8-Week course. I have removed all the inactive weeks from this repository to streamline it for convenient learning.