reinforcement learning (RL) has recently achieved tremendous successes in
many artificial intelligence applications. Many of the forefront applications
of RL involve multiple agents, e.g., playing chess and Go games, autonomous
driving, and robotics. Unfortunately, the framework upon w