Guardado en:
Detalles Bibliográficos
Autores principales: Chraibi, S., Iutzeler, F., Malick, J., Rogozin, A.
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2404.17190
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913330504925184
author Chraibi, S.
Iutzeler, F.
Malick, J.
Rogozin, A.
author_facet Chraibi, S.
Iutzeler, F.
Malick, J.
Rogozin, A.
contents Many problems in machine learning write as the minimization of a sum of individual loss functions over the training examples. These functions are usually differentiable but, in some cases, their gradients are not Lipschitz continuous, which compromises the use of (proximal) gradient algorithms. Fortunately, changing the geometry and using Bregman divergences can alleviate this issue in several applications, such as for Poisson linear inverse problems.However, the Bregman operation makes the aggregation of several points and gradients more involved, hindering the distribution of computations for such problems. In this paper, we propose an asynchronous variant of the Bregman proximal-gradient method, able to adapt to any centralized computing system. In particular, we prove that the algorithm copes with arbitrarily long delays and we illustrate its behavior on distributed Poisson inverse problems.
format Preprint
id arxiv_https___arxiv_org_abs_2404_17190
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Delay-tolerant distributed Bregman proximal algorithms
Chraibi, S.
Iutzeler, F.
Malick, J.
Rogozin, A.
Optimization and Control
Many problems in machine learning write as the minimization of a sum of individual loss functions over the training examples. These functions are usually differentiable but, in some cases, their gradients are not Lipschitz continuous, which compromises the use of (proximal) gradient algorithms. Fortunately, changing the geometry and using Bregman divergences can alleviate this issue in several applications, such as for Poisson linear inverse problems.However, the Bregman operation makes the aggregation of several points and gradients more involved, hindering the distribution of computations for such problems. In this paper, we propose an asynchronous variant of the Bregman proximal-gradient method, able to adapt to any centralized computing system. In particular, we prove that the algorithm copes with arbitrarily long delays and we illustrate its behavior on distributed Poisson inverse problems.
title Delay-tolerant distributed Bregman proximal algorithms
topic Optimization and Control
url https://arxiv.org/abs/2404.17190