arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2407.17815cs.GTcs.LG
收藏

嵌套复制者动力学、嵌套logit选择与基于相似性的学习

Nested replicator dynamics, nested logit choice, and similarity-based learning

Univ. Grenoble Alpes, CNRS, Inria, Grenoble INP, LIG(格勒诺布尔大学、法国国家科学研究中心、法国国家信息与自动化研究所、格勒诺布尔理工学院、信息与信号处理实验室);University of Wisconsin(威斯康星大学)

Panayotis Mertikopoulos, William H. Sandholm

AI总结:

本文提出嵌套复制者动力学,研究具有策略相似性结构的博弈学习模型,证明其虽不满足标准单调性假设但保留长期理性属性,并与嵌套logit选择规则及在线学习算法建立联系。

摘要:

我们考虑了博弈中的一种学习与演化模型,其行动集被赋予了一种基于划分的相似性结构,旨在捕捉策略间外生的相似性。在该模型中,修正策略的智能体有更高的概率将其当前策略与它们认为相似的其他策略进行比较,并且以与收益超额成正比的概率切换到所观察到的策略。由于这种对相似策略的隐性偏好,所得到的动力学——我们称之为嵌套复制者动力学——不满足模仿博弈动力学的任何标准单调性假设;然而,我们表明它们保留了复制者动力学的主要长期理性属性,尽管在量化速率上有所不同。我们还表明,所诱导的动力学可以被视为具有 Erev & Roth (1998) 精神的刺激-反应模型,其选择概率由 Ben-Akiva (1973) 和 McFadden (1978) 的嵌套logit选择规则给出。这一结果推广了在线学习中复制者动力学与指数权重算法之间的现有关系,并为我们的分析和结果提供了额外的解释层面。

英文摘要:

We consider a model of learning and evolution in games whose action sets are endowed with a partition-based similarity structure intended to capture exogenous similarities between strategies. In this model, revising agents have a higher probability of comparing their current strategy with other strategies that they deem similar, and they switch to the observed strategy with probability proportional to its payoff excess. Because of this implicit bias toward similar strategies, the resulting dynamics - which we call the nested replicator dynamics - do not satisfy any of the standard monotonicity postulates for imitative game dynamics; nonetheless, we show that they retain the main long-run rationality properties of the replicator dynamics, albeit at quantitatively different rates. We also show that the induced dynamics can be viewed as a stimulus-response model in the spirit of Erev & Roth (1998), with choice probabilities given by the nested logit choice rule of Ben-Akiva (1973) and McFadden (1978). This result generalizes an existing relation between the replicator dynamics and the exponential weights algorithm in online learning, and provides an additional layer of interpretation to our analysis and results.

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。

↑