Stochastic Optimization with Variance Reduction for Infinite Datasets with Finite-Sum Structure - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2017

Stochastic Optimization with Variance Reduction for Infinite Datasets with Finite-Sum Structure

Abstract

Stochastic optimization algorithms with variance reduction have proven successful for minimizing large finite sums of functions. Unfortunately, these techniques are unable to deal with stochastic perturbations of input data, induced for example by data augmentation. In such cases, the objective is no longer a finite sum, and the main candidate for optimization is the stochastic gradient descent method (SGD). In this paper, we introduce a variance reduction approach for these settings when the objective is composite and strongly convex. The convergence rate outperforms SGD with a typically much smaller constant factor, which depends on the variance of gradient estimates only due to perturbations on a single example.
Fichier principal
Vignette du fichier
stoch-miso-nips.pdf (580.18 Ko) Télécharger le fichier

Dates and versions

hal-01375816 , version 1 (03-10-2016)
hal-01375816 , version 2 (10-01-2017)
hal-01375816 , version 3 (23-01-2017)
hal-01375816 , version 4 (27-02-2017)
hal-01375816 , version 5 (01-06-2017)
hal-01375816 , version 6 (15-11-2017)

Identifiers

Cite

Alberto Bietti, Julien Mairal. Stochastic Optimization with Variance Reduction for Infinite Datasets with Finite-Sum Structure. NIPS 2017 - Advances in Neural Information Processing Systems, Dec 2017, Long Beach, CA, United States. pp.1622-1632. ⟨hal-01375816v6⟩
1180 View
756 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More