JaxAHT：基于JAX的临时团队协作库
JaxAHT: A JAX-Based Library for Ad Hoc Teamwork
浏览论文内容
中文总结 AI 辅助
JaxAHT是基于JAX的开源库，通过硬件加速和并行化统一AHT研究流程，实现约95倍加速，并提供多领域评估队友套件及基准研究，发现无算法普遍最优，建模在角色型场景中有效。
中文摘要 AI 辅助
临时团队协作（AHT）旨在解决设计能够与陌生伙伴协调而无需事先协调的智能体的挑战。然而，该领域的进展受到AHT研究生命周期计算成本过高、缺乏标准化基准实现以及缺少多样化且经过验证的评估队友套件的阻碍。在这项工作中，我们介绍了JaxAHT，这是首个基于JAX的开源库，旨在加速并标准化AHT研究生命周期。利用JAX的硬件加速和大规模并行化能力，JaxAHT为队友生成、自我智能体训练以及针对未见队友的评估提供了统一框架，相比PyTorch对应实现实现了约95倍的墙钟加速。除该库外，我们还贡献了一套涵盖基于级别的觅食、Overcooked和Hanabi领域的多样化评估队友。为展示该框架的价值，我们使用它进行了一项大规模、计算受控的基准研究，比较了队友生成和AHT智能体学习方法，发现没有算法始终表现最佳，且智能体建模主要在有不同队友的角色型场景中带来益处。
英文摘要
Ad Hoc Teamwork (AHT) addresses the challenge of designing agents capable of coordinating with novel partners without prior coordination. However, progress in the field is hindered by the prohibitive computational cost of the AHT research lifecycle, the lack of standardized benchmark implementations, and the absence of a diverse, validated evaluation teammate suite. In this work, we introduce JaxAHT, the first open-source, JAX-based library designed to accelerate and standardize the AHT research lifecycle. Leveraging JAX's hardware acceleration and massive parallelization capabilities, JaxAHT provides a unified framework for teammate generation, ego agent training, and evaluation against unseen teammates, achieving approximately 95x wall-clock speedup over PyTorch counterparts. Alongside the library, we contribute a diverse suite of evaluation teammates across the domains of Level-Based Foraging, Overcooked, and Hanabi. To illustrate the value of the framework, we use it to conduct a large-scale, compute-controlled benchmark study comparing teammate generation and AHT agent learning methods, finding that no algorithm consistently performs best, and that agent modeling primarily offers benefits in role-based scenarios with diverse teammates.
发表机构
- The University of Texas at Austin（德克萨斯大学奥斯汀分校）
- Google（谷歌）
- Sony AI（索尼人工智能）
机构由 AI 辅助整理，请以论文原文为准。
