arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2511.05005cs.LGcs.AIcs.RO
收藏

通过流匹配实现多智能体协调

Multi-agent Coordination via Flow Matching

University of Texas at Austin(德克萨斯大学奥斯汀分校);Sungkyunkwan University(松均大学)

Dongsu Lee, Daehee Lee, Amy Zhang

AI总结:

MAC-Flow通过流匹配方法实现多智能体协调,平衡了性能与计算效率,显著提升推理速度的同时保持良好表现。

摘要:

本文提出了MAC-Flow,一种简单而表达力强的多智能体协调框架。我们认为有效协调的要求是双重的:(i)对离线数据中多样化的联合行为有丰富的表示;(ii)在实时情况下能够高效行动。然而,先前的方法往往牺牲其中一方面,即去噪扩散基解决方案能够捕捉复杂的协调但计算速度慢,而高斯策略基解决方案则速度快但处理多智能体交互时易脆。MAC-Flow通过首先学习联合行为的流表示,然后将其蒸馏为去中心化的单步策略,以在保持协调的同时实现快速执行。在四个不同的基准测试中,包括12个环境和34个数据集,MAC-Flow缓解了性能和计算成本之间的权衡,具体实现了比扩散基MARL方法快约14.5倍的推理速度,同时保持良好的性能。同时,其推理速度与先前高斯策略基离线多智能体强化学习(MARL)方法相似。

英文摘要:

This work presents MAC-Flow, a simple yet expressive framework for multi-agent coordination. We argue that requirements of effective coordination are twofold: (i) a rich representation of the diverse joint behaviors present in offline data and (ii) the ability to act efficiently in real time. However, prior approaches often sacrifice one for the other, i.e., denoising diffusion-based solutions capture complex coordination but are computationally slow, while Gaussian policy-based solutions are fast but brittle in handling multi-agent interaction. MAC-Flow addresses this trade-off by first learning a flow-based representation of joint behaviors, and then distilling it into decentralized one-step policies that preserve coordination while enabling fast execution. Across four different benchmarks, including $12$ environments and $34$ datasets, MAC-Flow alleviates the trade-off between performance and computational cost, specifically achieving about $\boldsymbol{\times14.5}$ faster inference compared to diffusion-based MARL methods, while maintaining good performance. At the same time, its inference speed is similar to that of prior Gaussian policy-based offline multi-agent reinforcement learning (MARL) methods.

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。

↑