Do you want a RL agent nicely moving on Atari?
This is a step-by-step tutorial from DQN to Rainbow. Every chapter contains both of theoretical backgrounds and object-oriented implementation. Just pick any topic in which you are interested, and learn! You can execute them right away with Colab even on your smartphone.
Please feel free to open an issue or a pull-request if you have any idea to make it better. :)
If you want a tutorial for policy gradient methods, please see PG is All You Need.
- DQN [NBViewer] [Colab]
- DoubleDQN [NBViewer] [Colab]
- PrioritizedExperienceReplay [NBViewer] [Colab]
- DuelingNet [NBViewer] [Colab]
- NoisyNet [NBViewer] [Colab]
- CategoricalDQN [NBViewer] [Colab]
- N-stepLearning [NBViewer] [Colab]
- Rainbow [NBViewer] [Colab]
This repository is tested with python 3.8+
git clone https://github.com/Curt-Park/rainbow-is-all-you-need.git
cd rainbow-is-all-you-need
make setup
jupyter lab
- V. Mnih et al., "Human-level control through deep reinforcement learning." Nature, 518 (7540):529–533, 2015.
- van Hasselt et al., "Deep Reinforcement Learning with Double Q-learning." arXiv preprint arXiv:1509.06461, 2015.
- T. Schaul et al., "Prioritized Experience Replay." arXiv preprint arXiv:1511.05952, 2015.
- Z. Wang et al., "Dueling Network Architectures for Deep Reinforcement Learning." arXiv preprint arXiv:1511.06581, 2015.
- M. Fortunato et al., "Noisy Networks for Exploration." arXiv preprint arXiv:1706.10295, 2017.
- M. G. Bellemare et al., "A Distributional Perspective on Reinforcement Learning." arXiv preprint arXiv:1707.06887, 2017.
- R. S. Sutton, "Learning to predict by the methods of temporal differences." Machine learning, 3(1):9–44, 1988.
- M. Hessel et al., "Rainbow: Combining Improvements in Deep Reinforcement Learning." arXiv preprint arXiv:1710.02298, 2017.
Thanks goes to these wonderful people (emoji key):
Jinwoo Park (Curt) 💻 📖 |
Kyunghwan Kim 💻 |
Wei Chen 🚧 |
WANG Lei 🚧 |
leeyaf 💻 |
ahmadF 📖 |
Roberto Schiavone 💻 |
David Yuan 💻 |
dhanushka2001 💻 |
This project follows the all-contributors specification. Contributions of any kind welcome!