arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.06205cs.AIcs.GTcs.LGcs.MA
收藏

订单事项:代理人对代理人的策略优化

Order Matters: Agent-by-agent Policy Optimization

Shanghai Jiao Tong University(上海交通大学);Digital Brain Lab(数字大脑实验室);ShanghaiTech University(上海科技大学);University College London(伦敦大学学院)

Xihuai Wang, Zheng Tian, Ziyu Wan, Ying Wen, Jun Wang, Weinan Zhang

摘要:

虽然多智能体信赖域算法在解决协调任务方面取得了巨大的成功,但大多数算法都存在非平稳性问题,因为智能体会同时更新策略。相比之下,按代理更新策略的顺序方案提供了另一种视角,并显示出强大的性能。然而,样本效率低下和缺乏单调改善保证每个代理仍然是两个显着的挑战,顺序计划。在本文中,我们提出了\textbf{A} gent-by- \textbf{a} gent \textbf{P} policy \textbf{O} optimization(A2 PO)算法,以提高样本效率,并保留每个代理在训练过程中单调改进的保证。与其他信赖域算法相比,我们证明了单调改进界的紧密性。从顺序更新代理的角度出发,进一步考虑了代理更新顺序的影响,并将非平稳性理论推广到顺序更新方案中。为了评估A2 PO,我们对四个基准进行了全面的实证研究:StarCraftII,Multi-agent MuJoCo,Multi-agent Particle Environment和Google Research Football完整游戏场景。A2 PO的表现始终优于强基线。

英文摘要:

While multi-agent trust region algorithms have achieved great success empirically in solving coordination tasks, most of them, however, suffer from a non-stationarity problem since agents update their policies simultaneously. In contrast, a sequential scheme that updates policies agent-by-agent provides another perspective and shows strong performance. However, sample inefficiency and lack of monotonic improvement guarantees for each agent are still the two significant challenges for the sequential scheme. In this paper, we propose the \textbf{A}gent-by-\textbf{a}gent \textbf{P}olicy \textbf{O}ptimization (A2PO) algorithm to improve the sample efficiency and retain the guarantees of monotonic improvement for each agent during training. We justify the tightness of the monotonic improvement bound compared with other trust region algorithms. From the perspective of sequentially updating agents, we further consider the effect of agent updating order and extend the theory of non-stationarity into the sequential update scheme. To evaluate A2PO, we conduct a comprehensive empirical study on four benchmarks: StarCraftII, Multi-agent MuJoCo, Multi-agent Particle Environment, and Google Research Football full game scenarios. A2PO consistently outperforms strong baselines.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑