Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

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

A Learning Framework for Distribution-Based Game-Theoretic Solution Concepts The past few years have seen several works exploring learning economic solutions from data, including 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 solution concept can be efficiently 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 efficient solution learnability and show that results in existing works can be immediately derived using our methodology. Finally, we apply our methods in other economic domains, yielding learning variants of competitive equilibria and Condorcet winners. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png ACM Transactions on Economics and Computation Association for Computing Machinery

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

Download PDF
23 pages

Loading next page...
 
/lp/association-for-computing-machinery/a-learning-framework-for-distribution-based-game-theoretic-solution-6X6wMK00rF

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Association for Computing Machinery
Copyright
Copyright © 2023 Copyright held by the owner/author(s). Publication rights licensed to ACM.
ISSN
2167-8375
eISSN
2167-8383
DOI
10.1145/3580374
Publisher site
See Article on Publisher Site

Abstract

The past few years have seen several works exploring learning economic solutions from data, including 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 solution concept can be efficiently 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 efficient solution learnability and show that results in existing works can be immediately derived using our methodology. Finally, we apply our methods in other economic domains, yielding learning variants of competitive equilibria and Condorcet winners.

Journal

ACM Transactions on Economics and Computation – Association for Computing Machinery

Published: Jun 24, 2023

Keywords: Solution concepts

References

  • Neural Network Learning: Theoretical Foundations
    [,
  • Bounds on the cost of stabilizing a cooperative game
    [,
  • Learning cooperative games
    [,
  • Sample complexity of automated mechanism design
    [,
  • Envy-free classification
    [,
  • Learning submodular functions
    [,
  • Learning cooperative games
    [,
  • A general theory of sample complexity for multi-item profit maximization
    [,
  • The power of optimization from samples
    [,
  • The limitations of optimization from samples
    [,
  • The sample complexity of optimizing a convex function
    [,
  • Statistical cost sharing
    [,
  • Sample complexity for winner prediction in elections
    [,
  • Handbook of Computational Social Choice
    [,
  • Combinatorial auctions via machine learning-based preference elicitation
    [,
  • The combinatorial assignment problem: Approximate competitive equilibrium from equal incomes
    [,
  • The sample complexity of revenue maximization
    [,
  • Fixed point theorems for discontinuous mappings
    [,
  • Multiclass learnability and the ERM principle
    [,
  • The sample complexity of auctions with side information
    [,
  • Structured preferences
    [,
  • Walrasian equilibrium with gross substitutes
    [,
  • Deep learning for predicting human strategic behavior
    [,
  • Submodular optimization under noise
    [,
  • Hedonic games with graph-restricted communication
    [,
  • Forming probably stable communities with limited interactions
    [,
  • An Introduction to Computational Learning Theory
    [,
  • Learning cooperative solution concepts from voting behavior: A case study on the Israeli Knesset
    [,
  • The price is (probably) right: Learning market equilibria from samples
    [,
  • PrefLib: A library of preference data http://preflib
    [,
  • On the pseudo-dimension of nearly optimal auctions
    [,
  • Learning simple auctions
    [,
  • On learning sets and functions
    [,
  • Algorithmic Game Theory
    [,
  • The complexity of fairness through equilibrium
    [,
  • Learning to optimize combinatorial functions
    [,
  • Understanding Machine Learning: From Theory to Algorithms
    [,
  • Learnability, stability and uniform convergence
    [,
  • Learning hedonic games
    [,
  • A sample complexity measure with applications to learning optimal auctions
    [,
  • Chervonenkis
    [,
  • A PAC framework for aggregating agents’ judgments
    [,