[AAAI 2020] SMIX(λ): Enhancing Centralized Value Functions for Cooperative Multi-Agent Reinforcement Learning