Non-strongly-convex smooth stochastic approximation with convergence rate O(1/n) - Inria - Institut national de recherche en sciences et technologies du numérique
Conference Papers Year : 2013

Non-strongly-convex smooth stochastic approximation with convergence rate O(1/n)

Abstract

We consider the stochastic approximation problem where a convex function has to be minimized, given only the knowledge of unbiased estimates of its gradients at certain points, a framework which includes machine learning methods based on the minimization of the empirical risk. We focus on problems without strong convexity, for which all previously known algorithms achieve a convergence rate for function values of O(1/n^{1/2}). We consider and analyze two algorithms that achieve a rate of O(1/n) for classical supervised learning problems. For least-squares regression, we show that averaged stochastic gradient descent with constant step-size achieves the desired rate. For logistic regression, this is achieved by a simple novel stochastic gradient algorithm that (a) constructs successive local quadratic approximations of the loss functions, while (b) preserving the same running time complexity as stochastic gradient descent. For these algorithms, we provide a non-asymptotic analysis of the generalization error (in expectation, and also in high probability for least-squares), and run extensive experiments on standard machine learning benchmarks showing that they often outperform existing approaches.
Fichier principal
Vignette du fichier
newsto_hal.pdf (745.94 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-00831977 , version 1 (09-06-2013)

Identifiers

Cite

Francis Bach, Eric Moulines. Non-strongly-convex smooth stochastic approximation with convergence rate O(1/n). Neural Information Processing Systems (NIPS), Dec 2013, United States. ⟨hal-00831977⟩
705 View
1570 Download

Altmetric

Share

More