Problem-Parameter-Free Decentralized Nonconvex Stochastic Optimization

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Li, Jiaxiang, Chen, Xuxing, Ma, Shiqian, Hong, Mingyi
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914677925085184
author Li, Jiaxiang
Chen, Xuxing
Ma, Shiqian
Hong, Mingyi
author_facet Li, Jiaxiang
Chen, Xuxing
Ma, Shiqian
Hong, Mingyi
contents Existing decentralized algorithms usually require knowledge of problem parameters for updating local iterates. For example, the hyperparameters (such as learning rate) usually require the knowledge of Lipschitz constant of the global gradient or topological information of the communication networks, which are usually not accessible in practice. In this paper, we propose D-NASA, the first algorithm for decentralized nonconvex stochastic optimization that requires no prior knowledge of any problem parameters. We show that D-NASA has the optimal rate of convergence for nonconvex objectives under very mild conditions and enjoys the linear-speedup effect, i.e. the computation becomes faster as the number of nodes in the system increases. Extensive numerical experiments are conducted to support our findings.
format Preprint
id arxiv_https___arxiv_org_abs_2402_08821
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Problem-Parameter-Free Decentralized Nonconvex Stochastic Optimization
Li, Jiaxiang
Chen, Xuxing
Ma, Shiqian
Hong, Mingyi
Optimization and Control
Distributed, Parallel, and Cluster Computing
Existing decentralized algorithms usually require knowledge of problem parameters for updating local iterates. For example, the hyperparameters (such as learning rate) usually require the knowledge of Lipschitz constant of the global gradient or topological information of the communication networks, which are usually not accessible in practice. In this paper, we propose D-NASA, the first algorithm for decentralized nonconvex stochastic optimization that requires no prior knowledge of any problem parameters. We show that D-NASA has the optimal rate of convergence for nonconvex objectives under very mild conditions and enjoys the linear-speedup effect, i.e. the computation becomes faster as the number of nodes in the system increases. Extensive numerical experiments are conducted to support our findings.
title Problem-Parameter-Free Decentralized Nonconvex Stochastic Optimization
topic Optimization and Control
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2402.08821