arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.04831cs.AIcs.LG
收藏

面向零样本协作的开放式协作学习框架

Cooperative Open-ended Learning Framework for Zero-shot Coordination

The University of Manchester(曼彻斯特大学);Shanghai Jiao Tong University(上海交通大学);King’s College London(伦敦国王学院)

Yang Li, Shao Zhang, Jichen Sun, Yali Du, Ying Wen, Xinbing Wang, Wei Pan

AI总结:

针对合作AI零样本协作中的协作不相容问题,提出融合图论与博弈论的COLE开放式学习框架,并在Overcooked中验证其优于现有方法。

摘要:

合作式人工智能(AI)中的零样本协作仍是一项重大挑战,其含义是与各种从未见过的伙伴进行有效协作。以往算法试图通过在一个种群内部优化固定目标来提升策略或行为多样性,以应对这一挑战。然而,这些方法可能导致学习损失,并且无法与种群中的某些策略合作,这一问题被称为协作不相容性。为解决该问题,我们提出了 Cooperative Open-ended LEarning(COLE,开放式协作学习)框架。该框架从图论视角构建双人合作博弈中的开放式目标,以评估并识别每种策略的协作能力。我们进一步具体化该框架,并提出一种利用博弈论与图论知识的实用算法。此外,对该算法学习过程的分析表明,它能够有效克服协作不相容性。在 Overcooked 游戏环境中的实验结果表明,当与不同水平的伙伴协作时,我们的方法优于当前最先进的方法。演示可在 https://sites.google.com/view/cole-2023 获取。

英文摘要:

Zero-shot coordination in cooperative artificial intelligence (AI) remains a significant challenge, which means effectively coordinating with a wide range of unseen partners. Previous algorithms have attempted to address this challenge by optimizing fixed objectives within a population to improve strategy or behaviour diversity. However, these approaches can result in a loss of learning and an inability to cooperate with certain strategies within the population, known as cooperative incompatibility. To address this issue, we propose the Cooperative Open-ended LEarning (COLE) framework, which constructs open-ended objectives in cooperative games with two players from the perspective of graph theory to assess and identify the cooperative ability of each strategy. We further specify the framework and propose a practical algorithm that leverages knowledge from game theory and graph theory. Furthermore, an analysis of the learning process of the algorithm shows that it can efficiently overcome cooperative incompatibility. The experimental results in the Overcooked game environment demonstrate that our method outperforms current state-of-the-art methods when coordinating with different-level partners. Our demo is available at https://sites.google.com/view/cole-2023.

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。

↑