OPTIMAL NON-ASYMPTOTIC BOUND OF THE RUPPERT-POLYAK AVERAGING WITHOUT STRONG CONVEXITY - Laboratoire Angevin de Recherche en Mathématiques Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2017

OPTIMAL NON-ASYMPTOTIC BOUND OF THE RUPPERT-POLYAK AVERAGING WITHOUT STRONG CONVEXITY

Résumé

This paper is devoted to the non-asymptotic control of the mean-squared error for the Ruppert-Polyak stochastic averaged gradient descent introduced in the seminal contributions of [Rup88] and [PJ92]. In our main results, we establish non-asymptotic tight bounds (optimal with respect to the Cramer-Rao lower bound) in a very general framework that includes the uniformly strongly convex case as well as the one where the function f to be minimized satisfies a weaker Kurdyka-Lojiasewicz-type condition [Loj63, Kur98]. In particular, it makes it possible to recover some pathological examples such as on-line learning for logistic regression (see [Bac14]) and recursive quan-tile estimation (an even non-convex situation). Finally, our bound is optimal when the decreasing step (γn) n≥1 satisfies: γn = γn −β with β = 3/4, leading to a second-order term in O(n −5/4).
Fichier principal
Vignette du fichier
GP_09_20.pdf (519.09 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01623986 , version 1 (25-10-2017)

Identifiants

  • HAL Id : hal-01623986 , version 1

Citer

Sébastien Gadat, Fabien Panloup. OPTIMAL NON-ASYMPTOTIC BOUND OF THE RUPPERT-POLYAK AVERAGING WITHOUT STRONG CONVEXITY. 2017. ⟨hal-01623986⟩
264 Consultations
190 Téléchargements

Partager

Gmail Facebook X LinkedIn More