articleJan 1, 2013Closed access

Stochastic dual coordinate ascent methods for regularized loss

SSShai Shalev‐ShwartzTZTong Zhang

Hebrew University of Jerusalem · Rutgers, The State University of New Jersey

Abstract

Stochastic Gradient Descent (SGD) has become popular for solving large scale supervised machine learning optimization problems such as SVM, due to their strong theoretical guarantees. While the closely related Dual Coordinate Ascent (DCA) method has been implemented in various software pack-ages, it has so far lacked good convergence analysis. This paper presents a new analysis of Stochastic Dual Coordinate Ascent (SDCA) showing that this class of methods enjoy strong theoretical guarantees that are comparable or better than SGD. This analysis justifies the effectiveness of SDCA for practical applications. 1

Citation impact

699
total citations
FWCI
134.92
Percentile
100%
References
22
Citations per year

Authors

2

Topics & keywords

Topics
Keywords
  • Dual (grammatical number)
  • Computer science
  • Stochastic gradient descent
  • Convergence (economics)
  • Mathematical optimization
  • Coordinate descent
  • Support vector machine
  • Software
No related works found for this paper.