Continuous-Time Analysis of Heavy Ball Momentum in Min-Max Games

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Feng, Yi, Fujii, Kaito, Skoulakis, Stratis, Wang, Xiao, Cevher, Volkan
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912394860560384
author Feng, Yi
Fujii, Kaito
Skoulakis, Stratis
Wang, Xiao
Cevher, Volkan
author_facet Feng, Yi
Fujii, Kaito
Skoulakis, Stratis
Wang, Xiao
Cevher, Volkan
contents Since Polyak's pioneering work, heavy ball (HB) momentum has been widely studied in minimization. However, its role in min-max games remains largely unexplored. As a key component of practical min-max algorithms like Adam, this gap limits their effectiveness. In this paper, we present a continuous-time analysis for HB with simultaneous and alternating update schemes in min-max games. Locally, we prove smaller momentum enhances algorithmic stability by enabling local convergence across a wider range of step sizes, with alternating updates generally converging faster. Globally, we study the implicit regularization of HB, and find smaller momentum guides algorithms trajectories towards shallower slope regions of the loss landscapes, with alternating updates amplifying this effect. Surprisingly, all these phenomena differ from those observed in minimization, where larger momentum yields similar effects. Our results reveal fundamental differences between HB in min-max games and minimization, and numerical experiments further validate our theoretical results.
format Preprint
id arxiv_https___arxiv_org_abs_2505_19537
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Continuous-Time Analysis of Heavy Ball Momentum in Min-Max Games
Feng, Yi
Fujii, Kaito
Skoulakis, Stratis
Wang, Xiao
Cevher, Volkan
Computer Science and Game Theory
Machine Learning
Since Polyak's pioneering work, heavy ball (HB) momentum has been widely studied in minimization. However, its role in min-max games remains largely unexplored. As a key component of practical min-max algorithms like Adam, this gap limits their effectiveness. In this paper, we present a continuous-time analysis for HB with simultaneous and alternating update schemes in min-max games. Locally, we prove smaller momentum enhances algorithmic stability by enabling local convergence across a wider range of step sizes, with alternating updates generally converging faster. Globally, we study the implicit regularization of HB, and find smaller momentum guides algorithms trajectories towards shallower slope regions of the loss landscapes, with alternating updates amplifying this effect. Surprisingly, all these phenomena differ from those observed in minimization, where larger momentum yields similar effects. Our results reveal fundamental differences between HB in min-max games and minimization, and numerical experiments further validate our theoretical results.
title Continuous-Time Analysis of Heavy Ball Momentum in Min-Max Games
topic Computer Science and Game Theory
Machine Learning
url https://arxiv.org/abs/2505.19537