muzero-general and muzero
The kaesve implementation is a specialized fork/reimplementation of the werner-duvaud framework that adds MuZero-specific features (learned MDP models, inter-algorithm comparison) while maintaining compatibility with the AlphaZero General architecture, making them ecosystem variants rather than true competitors or complements.
About muzero-general
werner-duvaud/muzero-general
MuZero
This project offers a clear and documented re-implementation of the advanced MuZero AI algorithm, allowing you to train an AI to master games like Chess, Go, or Atari titles without explicit rules programming. You provide the game environment, and the AI learns to play and strategize, outputting a trained model capable of high-level play. It's designed for researchers, students, and enthusiasts interested in understanding and applying cutting-edge reinforcement learning.
About muzero
kaesve/muzero
A clean implementation of MuZero and AlphaZero following the AlphaZero General framework. Train and Pit both algorithms against each other, and investigate reliability of learned MuZero MDP models.
This project helps researchers and students understand and experiment with advanced AI decision-making algorithms like AlphaZero and MuZero. It takes configuration files defining AI agents and game environments, then trains the agents and allows them to compete. The output includes performance metrics and visualizations of how the AI models learn and make decisions, which is valuable for anyone studying or developing reinforcement learning.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work