A Class of Randomized Primal-Dual Algorithms for Distributed Optimization

Abstract : Based on a preconditioned version of the randomized block-coordinate forward-backward algorithm recently proposed in [23], several variants of block-coordinate primal-dual algo-rithms are designed in order to solve a wide array of monotone inclusion problems. These methods rely on a sweep of blocks of variables which are activated at each iteration according to a random rule, and they allow stochastic errors in the evaluation of the involved operators. Then, this framework is employed to derive block-coordinate primal-dual proximal algorithms for solving composite convex variational problems. The resulting algorithm implementations may be useful for reducing computational complexity and memory requirements. Furthermore, we show that the proposed approach can be used to develop novel asynchronous distributed primal-dual algorithms in a multi-agent context.
Type de document :
Pré-publication, Document de travail
Liste complète des métadonnées

Littérature citée [53 références]  Voir  Masquer  Télécharger

Contributeur : Audrey Repetti <>
Soumis le : dimanche 26 octobre 2014 - 00:05:11
Dernière modification le : jeudi 5 juillet 2018 - 14:46:11
Document(s) archivé(s) le : mardi 27 janvier 2015 - 10:05:37


Fichiers produits par l'(les) auteur(s)


  • HAL Id : hal-01077615, version 1
  • ARXIV : 1406.6404


Jean-Christophe Pesquet, Audrey Repetti. A Class of Randomized Primal-Dual Algorithms for Distributed Optimization. 2014. 〈hal-01077615〉



Consultations de la notice


Téléchargements de fichiers