Speed up reinforcement learning between two agents with adaptive mimetism | IEEE Conference Publication | IEEE Xplore