arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1903.08322cs.AIcs.GT
收藏

一种面向基于分布的博弈论解概念的学习框架

A Learning Framework for Distribution-Based Game-Theoretic Solution Concepts

International Institute of Information Technology, Hyderabad(海得拉巴国际信息技术学院);National University of Singapore(新加坡国立大学)

Tushant Jha, Yair Zick

AI总结:

本研究提出统一的学习理论框架,推广图维度概念以适配博弈论解概念学习,给出PAC可学习性充分条件,可推导现有结果并拓展至新经济领域提出新型解概念。

摘要:

过去几年间已有多项从数据中学习经济学解的研究,涵盖最优拍卖设计、函数优化、合作博弈中的稳定收益等诸多方向。本研究为这类问题的建模提供了一套统一的学习理论方法,并建立了用于判定给定经济学解概念能否从数据中学习的工具。我们的学习理论框架推广了函数空间维度的一种概念——图维度(graph dimension),使其适配解概念学习领域。我们明确了解概念具备PAC可学习性的充分条件,并证明现有研究中的结果可通过本方法直接推导得出。最后,我们将方法应用于其他经济学领域,提出了PAC竞争均衡与PAC孔多塞胜者(Condorcet winners)的全新概念。

英文摘要:

The past few years have seen several works on learning economic solutions from data; these include optimal auction design, function optimization, stable payoffs in cooperative games and more. In this work, we provide a unified learning-theoretic methodology for modeling such problems, and establish tools for determining whether a given economic solution concept can be learned from data. Our learning theoretic framework generalizes a notion of function space dimension -- the graph dimension -- adapting it to the solution concept learning domain. We identify sufficient conditions for the PAC learnability of solution concepts, and show that results in existing works can be immediately derived using our methodology. Finally, we apply our methods in other economic domains, yielding a novel notion of PAC competitive equilibrium and PAC Condorcet winners.

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。

↑