Independent Reinforcement Learning for Cooperative-Competitive Agents: Achieving Nash Equilibrium with Mean-Field Perspective

Developing a novel algorithm, MRPG, to achieve Nash equilibrium in cooperative-competitive multi-agent settings.