Learning Cooperative Multi-Agent Policies with Partial Reward Decoupling

no code implementations • 23 Dec 2021 • Benjamin Freed, Aditya Kapoor, Ian Abraham, Jeff Schneider, Howie Choset

One of the preeminent obstacles to scaling multi-agent reinforcement learning to large numbers of agents is assigning credit to individual agents' actions.

counterfactual Multi-agent Reinforcement Learning