研究方向:Cooperative Multi-Agent System base on Reinforcement Learning: Applying attention mechanism on multi-agent recurrent attention model, that each agent shares weighted feature collected from environment. Once features satisfied specific threshold, system then self-terminate with an action as output.