Code for a multi-agent particle environment used in the paper "Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments"
Why do you think that https://github.com/jindongwang/transferlearning is a good alternative to multiagent-particle-envs