Learning While Playing in Mean-Field Games: Convergence and Optimality

              Qiaomin Xie 1 Zhuoran Yang 2 Zhaoran Wang 3 Andreea Minca 1

             Abstract                 from the scalability issue. Specifically, in a multi-agent
                                  system, each agent interacts with the environment and the
   We study reinforcement learning in mean-field         other agents, with the goal of maximizing its own expected
   games. To achieve the Nash equilibrium, which ...