Summary This repository offers a curated collection of Multi-Agent Reinforcement Learning (MARL) resources, designed to accelerate beginners' understanding and learning process in the field by providing structured access to essential academic papers, practical tools, and research directions.

How It Works The project serves as an organized index of external MARL materials, not executable code. It categorizes learning pathways, including courses, conference proceedings (AAMAS, ICLR, etc.), review papers, books, and open-source environments (SMAC, EPyMARL), streamlining information discovery for MARL newcomers.

Quick Start & Requirements This repository is a collection of links and references. It contains no runnable code and thus has no installation or execution requirements.

Highlighted Details

• Resource Breadth: Features courses, major conferences (AAMAS, ICLR, etc.), review papers, books, and diverse open-source environments (SMAC, EPyMARL, MARLlib, PettingZoo).
• Ecosystem Mapping: Lists prominent research groups and companies in MARL, alongside curated paper lists and talks in English and Chinese.
• Key Environments: Highlights essential environments like StarCraft Micromanagement Environment (SMAC) and EPyMARL, recommended for beginners.
• Learning Aids: Includes links to academic databases (DBLP), code repositories (PapersWithCode), and educational platforms (OpenAI Spinning Up).

Maintenance & Community Described as a "first draft," the repository is intended for continuous updates and welcomes