Exploiting Approximate Symmetry for Efficient Multi-Agent Reinforcement Learning

Open in new window