Accelerated Proximal Stochastic Dual Coordinate Ascent for Regularized Loss Minimization
Machine Learning
2013-10-09 v2 Machine Learning
Numerical Analysis
Computation
Abstract
We introduce a proximal version of the stochastic dual coordinate ascent method and show how to accelerate the method using an inner-outer iteration procedure. We analyze the runtime of the framework and obtain rates that improve state-of-the-art results for various key machine learning optimization problems including SVM, logistic regression, ridge regression, Lasso, and multiclass SVM. Experiments validate our theoretical findings.
Cite
@article{arxiv.1309.2375,
title = {Accelerated Proximal Stochastic Dual Coordinate Ascent for Regularized Loss Minimization},
author = {Shai Shalev-Shwartz and Tong Zhang},
journal= {arXiv preprint arXiv:1309.2375},
year = {2013}
}