Fetching the paper…

RGMComm: Return Gap Minimization via Discrete Communications in Multi-Agent Reinforcement Learning · Around