Fast Last-Iterate Convergence of Learning in Games Requires Forgetful Algorithms

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cai, Yang, Farina, Gabriele, Grand-Clément, Julien, Kroer, Christian, Lee, Chung-Wei, Luo, Haipeng, Zheng, Weiqiang
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913657572556800
author Cai, Yang
Farina, Gabriele
Grand-Clément, Julien
Kroer, Christian
Lee, Chung-Wei
Luo, Haipeng
Zheng, Weiqiang
author_facet Cai, Yang
Farina, Gabriele
Grand-Clément, Julien
Kroer, Christian
Lee, Chung-Wei
Luo, Haipeng
Zheng, Weiqiang
contents Self-play via online learning is one of the premier ways to solve large-scale two-player zero-sum games, both in theory and practice. Particularly popular algorithms include optimistic multiplicative weights update (OMWU) and optimistic gradient-descent-ascent (OGDA). While both algorithms enjoy $O(1/T)$ ergodic convergence to Nash equilibrium in two-player zero-sum games, OMWU offers several advantages including logarithmic dependence on the size of the payoff matrix and $\widetilde{O}(1/T)$ convergence to coarse correlated equilibria even in general-sum games. However, in terms of last-iterate convergence in two-player zero-sum games, an increasingly popular topic in this area, OGDA guarantees that the duality gap shrinks at a rate of $O(1/\sqrt{T})$, while the best existing last-iterate convergence for OMWU depends on some game-dependent constant that could be arbitrarily large. This begs the question: is this potentially slow last-iterate convergence an inherent disadvantage of OMWU, or is the current analysis too loose? Somewhat surprisingly, we show that the former is true. More generally, we prove that a broad class of algorithms that do not forget the past quickly all suffer the same issue: for any arbitrarily small $δ>0$, there exists a $2\times 2$ matrix game such that the algorithm admits a constant duality gap even after $1/δ$ rounds. This class of algorithms includes OMWU and other standard optimistic follow-the-regularized-leader algorithms.
format Preprint
id arxiv_https___arxiv_org_abs_2406_10631
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Fast Last-Iterate Convergence of Learning in Games Requires Forgetful Algorithms
Cai, Yang
Farina, Gabriele
Grand-Clément, Julien
Kroer, Christian
Lee, Chung-Wei
Luo, Haipeng
Zheng, Weiqiang
Computer Science and Game Theory
Machine Learning
Optimization and Control
Self-play via online learning is one of the premier ways to solve large-scale two-player zero-sum games, both in theory and practice. Particularly popular algorithms include optimistic multiplicative weights update (OMWU) and optimistic gradient-descent-ascent (OGDA). While both algorithms enjoy $O(1/T)$ ergodic convergence to Nash equilibrium in two-player zero-sum games, OMWU offers several advantages including logarithmic dependence on the size of the payoff matrix and $\widetilde{O}(1/T)$ convergence to coarse correlated equilibria even in general-sum games. However, in terms of last-iterate convergence in two-player zero-sum games, an increasingly popular topic in this area, OGDA guarantees that the duality gap shrinks at a rate of $O(1/\sqrt{T})$, while the best existing last-iterate convergence for OMWU depends on some game-dependent constant that could be arbitrarily large. This begs the question: is this potentially slow last-iterate convergence an inherent disadvantage of OMWU, or is the current analysis too loose? Somewhat surprisingly, we show that the former is true. More generally, we prove that a broad class of algorithms that do not forget the past quickly all suffer the same issue: for any arbitrarily small $δ>0$, there exists a $2\times 2$ matrix game such that the algorithm admits a constant duality gap even after $1/δ$ rounds. This class of algorithms includes OMWU and other standard optimistic follow-the-regularized-leader algorithms.
title Fast Last-Iterate Convergence of Learning in Games Requires Forgetful Algorithms
topic Computer Science and Game Theory
Machine Learning
Optimization and Control
url https://arxiv.org/abs/2406.10631