Towards an optimal stochastic alternating direction method of multipliers

Samaneh Azadi, Suvrit Sra

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

22 Scopus citations

Abstract

We study regularized stochastic convex optimization subject to linear equality constraints. This class of problems was recently also studied by Ouyang et al. (2013) and Suzuki (2013); both introduced similar stochastic alternating direction method of multipliers (SADMM) algorithms. However, the analysis of both papers led to suboptimal convergence rates. This paper presents two new SADMM methods: (i) the first attains the minimax optimal rate of 0(l/κ2) for nonsmooth strongly-convex stochastic problems; while (ii) the second progresses towards an optimal rate by exhibiting an 0{I/κ2) rate for the smooth part. We present several experiments with our new methods; the results indicate improved performance over competing ADMM methods.

Original languageEnglish
Title of host publication31st International Conference on Machine Learning, ICML 2014
PublisherInternational Machine Learning Society (IMLS)
Pages944-959
Number of pages16
ISBN (Electronic)9781634393973
StatePublished - 2014
Externally publishedYes
Event31st International Conference on Machine Learning, ICML 2014 - Beijing, China
Duration: 21 Jun 201426 Jun 2014

Publication series

Name31st International Conference on Machine Learning, ICML 2014
Volume2

Conference

Conference31st International Conference on Machine Learning, ICML 2014
Country/TerritoryChina
CityBeijing
Period21/06/1426/06/14

Fingerprint

Dive into the research topics of 'Towards an optimal stochastic alternating direction method of multipliers'. Together they form a unique fingerprint.

Cite this