Multiagent reinforcement learning using function approximation | IEEE Journals & Magazine | IEEE Xplore