Local Steps Speed Up Local GD for Heterogeneous Distributed Logistic Regression

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Crawshaw, Michael, Woodworth, Blake, Liu, Mingrui
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910930710822912
author Crawshaw, Michael
Woodworth, Blake
Liu, Mingrui
author_facet Crawshaw, Michael
Woodworth, Blake
Liu, Mingrui
contents We analyze two variants of Local Gradient Descent applied to distributed logistic regression with heterogeneous, separable data and show convergence at the rate $O(1/KR)$ for $K$ local steps and sufficiently large $R$ communication rounds. In contrast, all existing convergence guarantees for Local GD applied to any problem are at least $Ω(1/R)$, meaning they fail to show the benefit of local updates. The key to our improved guarantee is showing progress on the logistic regression objective when using a large stepsize $η\gg 1/K$, whereas prior analysis depends on $η\leq 1/K$.
format Preprint
id arxiv_https___arxiv_org_abs_2501_13790
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Local Steps Speed Up Local GD for Heterogeneous Distributed Logistic Regression
Crawshaw, Michael
Woodworth, Blake
Liu, Mingrui
Machine Learning
We analyze two variants of Local Gradient Descent applied to distributed logistic regression with heterogeneous, separable data and show convergence at the rate $O(1/KR)$ for $K$ local steps and sufficiently large $R$ communication rounds. In contrast, all existing convergence guarantees for Local GD applied to any problem are at least $Ω(1/R)$, meaning they fail to show the benefit of local updates. The key to our improved guarantee is showing progress on the logistic regression objective when using a large stepsize $η\gg 1/K$, whereas prior analysis depends on $η\leq 1/K$.
title Local Steps Speed Up Local GD for Heterogeneous Distributed Logistic Regression
topic Machine Learning
url https://arxiv.org/abs/2501.13790