Saved in:
Bibliographic Details
Main Authors: Yoon, TaeHo, Kim, Jaeyeon, Suh, Jaewook J., Ryu, Ernest K.
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2404.13228
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917648407724032
author Yoon, TaeHo
Kim, Jaeyeon
Suh, Jaewook J.
Ryu, Ernest K.
author_facet Yoon, TaeHo
Kim, Jaeyeon
Suh, Jaewook J.
Ryu, Ernest K.
contents Recently, accelerated algorithms using the anchoring mechanism for minimax optimization and fixed-point problems have been proposed, and matching complexity lower bounds establish their optimality. In this work, we present the surprising observation that the optimal acceleration mechanism in minimax optimization and fixed-point problems is not unique. Our new algorithms achieve exactly the same worst-case convergence rates as existing anchor-based methods while using materially different acceleration mechanisms. Specifically, these new algorithms are dual to the prior anchor-based accelerated methods in the sense of H-duality. This finding opens a new avenue of research on accelerated algorithms since we now have a family of methods that empirically exhibit varied characteristics while having the same optimal worst-case guarantee.
format Preprint
id arxiv_https___arxiv_org_abs_2404_13228
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Optimal Acceleration for Minimax and Fixed-Point Problems is Not Unique
Yoon, TaeHo
Kim, Jaeyeon
Suh, Jaewook J.
Ryu, Ernest K.
Optimization and Control
Recently, accelerated algorithms using the anchoring mechanism for minimax optimization and fixed-point problems have been proposed, and matching complexity lower bounds establish their optimality. In this work, we present the surprising observation that the optimal acceleration mechanism in minimax optimization and fixed-point problems is not unique. Our new algorithms achieve exactly the same worst-case convergence rates as existing anchor-based methods while using materially different acceleration mechanisms. Specifically, these new algorithms are dual to the prior anchor-based accelerated methods in the sense of H-duality. This finding opens a new avenue of research on accelerated algorithms since we now have a family of methods that empirically exhibit varied characteristics while having the same optimal worst-case guarantee.
title Optimal Acceleration for Minimax and Fixed-Point Problems is Not Unique
topic Optimization and Control
url https://arxiv.org/abs/2404.13228