Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Bridging the Gap between Constant Step Size Stochastic Gradient Descent and Markov Chains

Aymeric Dieuleveut 1, 2 Alain Durmus 3 Francis Bach 1
1 SIERRA - Statistical Machine Learning and Parsimony
DI-ENS - Département d'informatique de l'École normale supérieure, CNRS - Centre National de la Recherche Scientifique, Inria de Paris
Abstract : We consider the minimization of an objective function given access to unbiased estimates of its gradient through stochastic gradient descent (SGD) with constant step-size. While the detailed analysis was only performed for quadratic functions, we provide an explicit asymptotic expansion of the moments of the averaged SGD iterates that outlines the dependence on initial conditions, the effect of noise and the step-size, as well as the lack of convergence in the general (non-quadratic) case. For this analysis, we bring tools from Markov chain theory into the analysis of stochastic gradient. We then show that Richardson-Romberg extrapolation may be used to get closer to the global optimum and we show empirical improvements of the new extrapolation scheme.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

Cited literature [59 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01565514
Contributor : Alain Durmus <>
Submitted on : Tuesday, April 10, 2018 - 9:09:38 PM
Last modification on : Tuesday, September 22, 2020 - 3:50:21 AM

Files

main_arxiv.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01565514, version 2
  • ARXIV : 1707.06386

Citation

Aymeric Dieuleveut, Alain Durmus, Francis Bach. Bridging the Gap between Constant Step Size Stochastic Gradient Descent and Markov Chains. 2018. ⟨hal-01565514v2⟩

Share

Metrics

Record views

581

Files downloads

1464