Computing Optimal Equilibria and Mechanisms via Learning in Zero-Sum Extensive-Form Games

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Brian Hu, Farina, Gabriele, Anagnostides, Ioannis, Cacciamani, Federico, McAleer, Stephen Marcus, Haupt, Andreas Alexander, Celli, Andrea, Gatti, Nicola, Conitzer, Vincent, Sandholm, Tuomas
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909209550913536
author Zhang, Brian Hu
Farina, Gabriele
Anagnostides, Ioannis
Cacciamani, Federico
McAleer, Stephen Marcus
Haupt, Andreas Alexander
Celli, Andrea
Gatti, Nicola
Conitzer, Vincent
Sandholm, Tuomas
author_facet Zhang, Brian Hu
Farina, Gabriele
Anagnostides, Ioannis
Cacciamani, Federico
McAleer, Stephen Marcus
Haupt, Andreas Alexander
Celli, Andrea
Gatti, Nicola
Conitzer, Vincent
Sandholm, Tuomas
contents We introduce a new approach for computing optimal equilibria via learning in games. It applies to extensive-form settings with any number of players, including mechanism design, information design, and solution concepts such as correlated, communication, and certification equilibria. We observe that optimal equilibria are minimax equilibrium strategies of a player in an extensive-form zero-sum game. This reformulation allows to apply techniques for learning in zero-sum games, yielding the first learning dynamics that converge to optimal equilibria, not only in empirical averages, but also in iterates. We demonstrate the practical scalability and flexibility of our approach by attaining state-of-the-art performance in benchmark tabular games, and by computing an optimal mechanism for a sequential auction design problem using deep reinforcement learning.
format Preprint
id arxiv_https___arxiv_org_abs_2306_05216
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Computing Optimal Equilibria and Mechanisms via Learning in Zero-Sum Extensive-Form Games
Zhang, Brian Hu
Farina, Gabriele
Anagnostides, Ioannis
Cacciamani, Federico
McAleer, Stephen Marcus
Haupt, Andreas Alexander
Celli, Andrea
Gatti, Nicola
Conitzer, Vincent
Sandholm, Tuomas
Computer Science and Game Theory
We introduce a new approach for computing optimal equilibria via learning in games. It applies to extensive-form settings with any number of players, including mechanism design, information design, and solution concepts such as correlated, communication, and certification equilibria. We observe that optimal equilibria are minimax equilibrium strategies of a player in an extensive-form zero-sum game. This reformulation allows to apply techniques for learning in zero-sum games, yielding the first learning dynamics that converge to optimal equilibria, not only in empirical averages, but also in iterates. We demonstrate the practical scalability and flexibility of our approach by attaining state-of-the-art performance in benchmark tabular games, and by computing an optimal mechanism for a sequential auction design problem using deep reinforcement learning.
title Computing Optimal Equilibria and Mechanisms via Learning in Zero-Sum Extensive-Form Games
topic Computer Science and Game Theory
url https://arxiv.org/abs/2306.05216