Online Asynchronous Distributed Regression - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Annales de l'ISUP Année : 2018

Online Asynchronous Distributed Regression

Résumé

Distributed computing offers a high degree of flexibility to accommodate modern learning constraints and the ever increasing size of datasets involved in massive data issues. Drawing inspiration from the theory of distributed computation models developed in the context of gradient-type optimization algorithms, we present a consensus-based asynchronous distributed approach for nonparametric online regression and analyze some of its asymptotic properties. Substantial numerical evidence involving up to 28 parallel processors is provided on synthetic datasets to assess the excellent performance of our method, both in terms of computation time and prediction accuracy.
Fichier principal
Vignette du fichier
Pages de DEP_8-V-64396_(2015-2019)-32 (1).pdf (11.5 Mo) Télécharger le fichier
Origine : Accord explicite pour ce dépôt

Dates et versions

hal-01024673 , version 1 (16-07-2014)
hal-01024673 , version 2 (10-03-2022)

Identifiants

Citer

Gérard Biau, Ryad Zenine. Online Asynchronous Distributed Regression. Annales de l'ISUP, 2018, 62 (3), pp.29-58. ⟨hal-01024673v2⟩
440 Consultations
151 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More