Salvato in:
Dettagli Bibliografici
Autori principali: Malitsky, Yura, Tam, Matthew K.
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:https://arxiv.org/abs/2308.11876
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914261547089920
author Malitsky, Yura
Tam, Matthew K.
author_facet Malitsky, Yura
Tam, Matthew K.
contents In this work, we consider a connected network of finitely many agents working cooperatively to solve a min-max problem with convex-concave structure. We propose a decentralised first-order algorithm which can be viewed as a non-trivial combination of two algorithms: PG-EXTRA for decentralised minimisation problems and the forward reflected backward method for (non-distributed) min-max problems. In each iteration of our algorithm, each agent computes the gradient of the smooth component of its local objective function as well as the proximal operator of its nonsmooth component, following by a round of communication with its neighbours. Our analysis shows that the sequence generated by the method converges under standard assumptions with non-decaying stepsize.
format Preprint
id arxiv_https___arxiv_org_abs_2308_11876
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle A First-Order Algorithm for Decentralised Min-Max Problems
Malitsky, Yura
Tam, Matthew K.
Optimization and Control
Distributed, Parallel, and Cluster Computing
In this work, we consider a connected network of finitely many agents working cooperatively to solve a min-max problem with convex-concave structure. We propose a decentralised first-order algorithm which can be viewed as a non-trivial combination of two algorithms: PG-EXTRA for decentralised minimisation problems and the forward reflected backward method for (non-distributed) min-max problems. In each iteration of our algorithm, each agent computes the gradient of the smooth component of its local objective function as well as the proximal operator of its nonsmooth component, following by a round of communication with its neighbours. Our analysis shows that the sequence generated by the method converges under standard assumptions with non-decaying stepsize.
title A First-Order Algorithm for Decentralised Min-Max Problems
topic Optimization and Control
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2308.11876