jianzhnie / RLZero

A clean and easy implementation of MuZero, AlphaZero and Self-Play reinforcement learning algorithms for any game.

RLZero

A clean and easy implementation of MuZero, AlphaZero and Self-Play reinforcement learning algorithms for any game.

References

Unified Toolkits

• https://github.com/datamllab/rlcard
• https://github.com/google-deepmind/open_spiel
• https://github.com/sotetsuk/pgx.git
• https://github.com/Unity-Technologies/ml-agents
• https://github.com/opendilab/LightZero

MCTS

https://github.com/masouduut94/MCTS-agent-python

AlpahZero

• https://github.com/suragnair/alpha-zero-general
• https://github.com/geochri/AlphaZero_Chess
• https://github.com/junxiaosong/AlphaZero_Gomoku
• https://github.com/lowrollr/turbozero
• https://github.com/dylandjian/SuperGo

MuZero

• https://github.com/werner-duvaud/muzero-general
• https://github.com/koulanurag/muzero-pytorch
• https://github.com/YeWR/EfficientZero.git

DouZero

• https://github.com/kwai/DouZero.git

Blog

• https://medium.com/@bentou.pub/

• https://github.com/BentouAI/AlphaZero-Chain-Reaction

• https://medium.com/applied-data-science/alphago-zero-explained-in-one-diagram-365f5abf67e0

• https://bbs.huaweicloud.com/blogs/384922

• http://xtf615.com/categories/%E5%BC%BA%E5%8C%96%E5%AD%A6%E4%B9%A0/

• How to Build Your Own MuZero Using Python (Part 1/3)

• How to Build Your Own MuZero Using Python (Part 2/3)

• How to Build Your Own MuZero Using Python (Part 3/3)

• 使用AlphaZero算法打造属于你自己的象棋AI!

• 用飞桨框架2.0造一个会下五子棋的AI模型

• AlphaZero实现Connect4游戏

• 极简MuZero算法实践——Paddle2.0版本

MCTS 的前沿动态

如何追踪 MCTS 的前沿动态？（1） 如何追踪 MCTS 的前沿动态？（2） 如何追踪 MCTS 的前沿动态？（3） 如何追踪 MCTS 的前沿动态？（4）

---

MCTS + RL 系列技术博客（1）：AlphaZero MCTS + RL 系列技术博客（2）：MuZero MCTS + RL 系列技术博客（3）：Sampled MuZero MCTS + RL 系列技术博客（4）：EfficientZero MCTS + RL 系列技术博客（5）：Stochastic MuZero MCTS + RL 系列技术博客（6）：浅析 MCTS 算法原理演进史