Changelog:
+ October 22, 2022: Upload MARL results!

Benchmark result of different RL algorithms on MetaDrive environments, including Multi-agent RL (IPPO, centralized critics, CoPO).

Please setup via:

git clone https://github.com/metadriverse/metadrive-benchmark.git
cd metadrive-benchmark

pip install matplotlib seaborn jupyter notebook

You can find formal experiment results in several MetaDrive environments in corresponding folders:

• metadrive-benchmark/MARL - Multi-agent Reinforcement Learning

Please cite us via: https://github.com/metadriverse/metadrive#-references