Saved in:
Bibliographic Details
Main Authors: Jha, Tushant, Zick, Yair
Format: Preprint
Published: 2019
Subjects:
Online Access:https://arxiv.org/abs/1903.08322
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918022301614080
author Jha, Tushant
Zick, Yair
author_facet Jha, Tushant
Zick, Yair
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_1903_08322
institution arXiv
publishDate 2019
record_format arxiv
spellingShingle A Learning Framework for Distribution-Based Game-Theoretic Solution Concepts
Jha, Tushant
Zick, Yair
Artificial Intelligence
Computer Science and Game Theory
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.
title A Learning Framework for Distribution-Based Game-Theoretic Solution Concepts
topic Artificial Intelligence
Computer Science and Game Theory
url https://arxiv.org/abs/1903.08322