Reinforcement-Learning
A repo of deep reinforcement learning projects
Key algorithms to implement
- DQN - DONE
- Tutorial on Cartpole: https://pytorch.org/tutorials/intermediate/reinforcement_q_learning.html (can extend to atari games)
- Policy Gradient (REINFORCE)
- A2C
- A3C
- (Optional) TRPO
- PPO (important)
- DDPG
- Twin Delayed DDPG
- Soft Actor-Critic
Sutton Textbook solutions
https://micahcarroll.github.io/learning/2018/05/17/sutton-and-barto-rl.html
Openai spinup
https://spinningup.openai.com/en/latest/index.html
- Introduces key SOTA algorithms to learn
- TODO: find pytorch implementations of listed algorithms
Examples using spinup models
https://github.com/openai/spinningup/tree/master/spinup/examples
https://spinningup.openai.com/en/latest/user/running.html#launching-from-scripts
Pytorch baseline models (best for implementation of RL algorithms)
https://github.com/zuoxingdong/lagom
https://github.com/openai/baselines
https://github.com/hill-a/stable-baselines
videos
https://www.youtube.com/channel/UCTgM-VlXKuylPrZ_YGAJHOw
https://www.youtube.com/channel/UC7ZVvEo7-B7lA6LY2MVX72A
Pytorch resources
https://github.com/qfettes/DeepRL-Tutorials (Done on Atari- Pong)
https://github.com/yandexdataschool/Practical_RL (check out some of their homework assignments)
https://github.com/ikostrikov/pytorch-rl
https://github.com/ikostrikov?tab=repositories
https://github.com/vitchyr/rlkit
Pytorch Atari with A3C lstm
https://github.com/Nasdin/ReinforcementLearning-AtariGame
Useful links
Start with this one https://github.com/sudharsan13296/Hands-On-Reinforcement-Learning-With-Python
https://github.com/MorvanZhou/Reinforcement-learning-with-tensorflow
https://github.com/simoninithomas/Deep_reinforcement_learning_Course
https://github.com/dennybritz/reinforcement-learning
A2C in TF2: http://inoryy.com/post/tensorflow2-deep-reinforcement-learning/