magent 0.1.3

Multi-Agent Reinforcement Learning environments with very large numbers of agents

MAgent

MAgent is a library for creating 2D environments with very large numbers of agents for conducting research in Multi-Agent Reinforcement Learning. These can look like this:

This is a maintained fork from the original repo- https://github.com/geek-ai/MAgent.

Requirements

MAgent supports Linux and OS X and Python 3.5+

Install instructions

Linux:

sudo apt-get install cmake libboost-system-dev libjsoncpp-dev libwebsocketpp-dev

pip3 install magent

macOS:

brew install cmake llvm boost@1.55
brew install jsoncpp argp-standalone
brew tap davidzhen0/homebrew-websocketpp
brew install --HEAD davidzhen0/websocketpp/websocketpp
brew link --force boost@1.55

pip3 install magent

If you use this in your research, please cite the original paper:

@inproceedings{zheng2018magent,
  title={MAgent: A many-agent reinforcement learning platform for artificial collective intelligence},
  author={Zheng, Lianmin and Yang, Jiacheng and Cai, Han and Zhou, Ming and Zhang, Weinan and Wang, Jun and Yu, Yong},
  booktitle={Thirty-Second AAAI Conference on Artificial Intelligence},
  year={2018}
}