Offline Multi-Agent Reinforcement Learning

Offline reinforcement learning leverages previously collected offline datasets to learn optimal policies with no necessity to access …