This was part of Advances in Optimal Decision Making under Uncertainty

Mean-Field Multi-Agent Reinforcement Learning: A Decentralized Network Approach

Xin Guo, University of California, Berkeley