提出N-XPlay框架，实现多团队间零样本协作
Towards Zero-Shot Coordination between Teams of Agents: The N-XPlay Framework
- 设计N人版本Overcooked，支持多智能体协作评估
- 在2/3/5人场景中，跨团队协作表现优于自对弈训练
- 适合研究多团队协同的智能体系统开发者
零样本协作（ZSC）——与陌生伙伴协作的能力——是使自主智能体成为有效队友的关键。现有ZSC方法仅评估两个未交互过的智能体间的协作能力，但无法反映真实多智能体系统中的复杂性，后者常涉及子群体层级结构及多团队间的互动，即多团队系统（MTS）。为此，我们首先引入N-player Overcooked，作为流行双智能体ZSC基准的扩展，支持在多智能体场景下评估ZSC。随后提出N-XPlay框架，用于解决多智能体、多团队设置下的零样本协作问题。在两、三、五人Overcooked场景中，将智能体分为“自我团队”和未知合作者组进行对比实验，结果显示，采用N-XPlay训练的智能体在同时平衡“团队内”与“团队间”协作方面，优于自对弈（Self-Play, SP）训练的智能体。
原文摘要 · Abstract (English)
Zero-shot coordination (ZSC) -- the ability to collaborate with unfamiliar partners -- is essential to making autonomous agents effective teammates. Existing ZSC methods evaluate coordination capabilities between two agents who have not previously interacted. However, these scenarios do not reflect the complexity of real-world multi-agent systems, where coordination often involves a hierarchy of sub-groups and interactions between teams of agents, known as Multi-Team Systems (MTS). To address this gap, we first introduce N-player Overcooked, an N-agent extension of the popular two-agent ZSC benchmark, enabling evaluation of ZSC in N-agent scenarios. We then propose N-XPlay for ZSC in N-agent, multi-team settings. Comparison against Self-Play across two-, three- and five-player Overcooked scenarios, where agents are split between an ``ego-team'' and a group of unseen collaborators shows that agents trained with N-XPlay are better able to simultaneously balance ``intra-team'' and ``inter-team'' coordination than agents trained with SP.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。