arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2201.06257cs.MA
收藏

GCS:基于图的多智能体强化学习协调策略

GCS: Graph-based Coordination Strategy for Multi-Agent Reinforcement Learning

Jingqing Ruan, Yali Du, Xuantang Xiong, Dengpeng Xing, Xiyun Li, Linghui Meng, Haifeng Zhang, Jun Wang, Bo Xu

摘要:

许多真实世界的场景涉及一组代理,他们必须协调他们的策略以实现共同的目标。以往的研究主要集中在分散控制上,以最大化共同的报酬,而很少考虑控制策略之间的协调,这在动态复杂环境中是至关重要的。在这项工作中,我们建议将联合团队策略分解为图形生成器和基于图形的协调策略,以实现代理之间的协调行为。图形生成器采用编码器-解码器框架,输出有向无环图(DAG)来捕获底层动态决策结构。我们还在图生成器中应用了DAGness-constrained和DAG depth-constrained优化来平衡效率和性能。基于图的协调策略利用生成的决策结构。同时训练图生成器和协调策略以最大化贴现收益。对协同高斯挤压、协同导航和Google Research Football的实验结果表明了该方法的优越性。

英文摘要:

Many real-world scenarios involve a team of agents that have to coordinate their policies to achieve a shared goal. Previous studies mainly focus on decentralized control to maximize a common reward and barely consider the coordination among control policies, which is critical in dynamic and complicated environments. In this work, we propose factorizing the joint team policy into a graph generator and graph-based coordinated policy to enable coordinated behaviours among agents. The graph generator adopts an encoder-decoder framework that outputs directed acyclic graphs (DAGs) to capture the underlying dynamic decision structure. We also apply the DAGness-constrained and DAG depth-constrained optimization in the graph generator to balance efficiency and performance. The graph-based coordinated policy exploits the generated decision structure. The graph generator and coordinated policy are trained simultaneously to maximize the discounted return. Empirical evaluations on Collaborative Gaussian Squeeze, Cooperative Navigation, and Google Research Football demonstrate the superiority of the proposed method.

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。

↑