STAS: 多智能体强化学习的时空回报分解（英文版）

英文标题：STAS: Spatial-Temporal Return Decomposition for Multi-agent Reinforcement Learning中文摘要：提出了一种名为 Spatial-Temporal Attention with Shapley（STAS）的新方法，该方法可以在时间和空间维度上学习信用分配，在多智能体强化学习中实现有效的空间 - 时间信用分配，优于所有现有的基线。英文摘要：Centralized Training with Decentralized Execution (CTDE) has been proven tobe an effective paradigm in cooperative multi-agent reinforcement learning(MARL). One of the major challenges is yet credit assignment, which aims tocredit agents by their contributions. Prior studies focus on e