Mini-batch primal and dual methods for SVMs

Martin Takáč, Avleen Bijral, Peter Richtarik, Nathan Srebro

Research output: Contribution to conferencePaperpeer-review

58 Scopus citations

Abstract

We address the issue of using mini-batches in stochastic optimization of SVMs. We show that the same quantity, the spectral norm of the data, controls the parallelization speedup obtained for both primal stochastic subgradient descent (SGD) and stochastic dual coordinate ascent (SCDA) methods and use it to derive novel variants of mini-batched SDCA. Our guarantees for both methods are expressed in terms of the original nonsmooth primal problem based on the hinge-loss.

Original languageEnglish (US)
Pages2059-2067
Number of pages9
StatePublished - Jan 1 2013
Event30th International Conference on Machine Learning, ICML 2013 - Atlanta, GA, United States
Duration: Jun 16 2013Jun 21 2013

Other

Other30th International Conference on Machine Learning, ICML 2013
CountryUnited States
CityAtlanta, GA
Period06/16/1306/21/13

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Sociology and Political Science

Fingerprint

Dive into the research topics of 'Mini-batch primal and dual methods for SVMs'. Together they form a unique fingerprint.

Cite this