Inexact and Stochastic Generalized Conditional Gradient with Augmented Lagrangian and Proximal Step - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Journal of Nonsmooth Analysis and Optimization Année : 2021

Inexact and Stochastic Generalized Conditional Gradient with Augmented Lagrangian and Proximal Step

Résumé

In this paper we propose and analyze inexact and stochastic versions of the CGALP algorithm developed in [25], which we denote ICGALP , that allow for errors in the computation of several important quantities. In particular this allows one to compute some gradients, proximal terms, and/or linear minimization oracles in an inexact fashion that facilitates the practical application of the algorithm to computationally intensive settings, e.g., in high (or possibly infinite) dimensional Hilbert spaces commonly found in machine learning problems. The algorithm is able to solve composite minimization problems involving the sum of three convex proper lower-semicontinuous functions subject to an affine constraint of the form Ax = b for some bounded linear operator A. Only one of the functions in the objective is assumed to be differentiable, the other two are assumed to have an accessible proximal operator and a linear minimization oracle. As main results, we show convergence of the Lagrangian values (so-called convergence in the Bregman sense) and asymptotic feasibility of the affine constraint as well as strong convergence of the sequence of dual variables to a solution of the dual problem, in an almost sure sense. Almost sure convergence rates are given for the Lagrangian values and the feasibility gap for the ergodic primal variables. Rates in expectation are given for the Lagrangian values and the feasibility gap subsequentially in the pointwise sense. Numerical experiments verifying the predicted rates of convergence are shown as well.
Fichier principal
Vignette du fichier
cgalp_inexact_1.pdf (780.62 Ko) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte

Dates et versions

hal-02569925 , version 1 (11-05-2020)
hal-02569925 , version 2 (05-01-2021)
hal-02569925 , version 3 (28-05-2021)
hal-02569925 , version 4 (31-08-2021)

Identifiants

Citer

Antonio Silveti-Falls, Cesare Molinari, Jalal Fadili. Inexact and Stochastic Generalized Conditional Gradient with Augmented Lagrangian and Proximal Step. Journal of Nonsmooth Analysis and Optimization, 2021, Volume 2 (6480), pp.41. ⟨10.46298/jnsao-2021-6480⟩. ⟨hal-02569925v4⟩
120 Consultations
556 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More