This repository is a curated list of resources for Game AI, specifically focusing on multi-agent reinforcement learning in both perfect and imperfect information games. It targets researchers and developers in the field of AI for games, providing a structured overview of open-source projects, papers, conferences, and competitions to accelerate learning and development.

How It Works

The collection categorizes resources by game (e.g., Texas Hold'em, Dou Dizhu, Starcraft, Go) and sorts papers by year. It highlights key areas like unified toolkits (RLCard, OpenSpiel, Unity ML-Agents), specific game implementations, and foundational research papers in multi-agent reinforcement learning, including seminal works like AlphaGo and AlphaZero.

Quick Start & Requirements

This is a curated list, not a runnable project. Links to code repositories and papers are provided for individual projects.

Highlighted Details

• Comprehensive coverage of major games like Poker, Dou Dizhu, Starcraft, Go, and Chess.
• Includes links to both foundational and recent research papers, often with associated code.
• Features unified toolkits for reinforcement learning in games, facilitating experimentation.
• Organizes resources by game type and publication year for easy navigation.

Maintenance & Community

The list is maintained by datamllab. Contributions are welcomed via pull requests. Contact information for maintainers is provided.

Licensing & Compatibility

Resource licensing varies by individual project. The repository itself is not licensed as a software project.

Limitations & Caveats

The README explicitly states the list is "incomplete." While extensive, it may not cover all available resources in the rapidly evolving field of game AI.