English

A Quasi-Newton Method for Large Scale Support Vector Machines

Machine Learning 2014-02-21 v1

Abstract

This paper adapts a recently developed regularized stochastic version of the Broyden, Fletcher, Goldfarb, and Shanno (BFGS) quasi-Newton method for the solution of support vector machine classification problems. The proposed method is shown to converge almost surely to the optimal classifier at a rate that is linear in expectation. Numerical results show that the proposed method exhibits a convergence rate that degrades smoothly with the dimensionality of the feature vectors.

Keywords

Cite

@article{arxiv.1402.4861,
  title  = {A Quasi-Newton Method for Large Scale Support Vector Machines},
  author = {Aryan Mokhtari and Alejandro Ribeiro},
  journal= {arXiv preprint arXiv:1402.4861},
  year   = {2014}
}

Comments

5 pages, To appear in International Conference on Acoustics, Speech, and Signal Processing (ICASSP) 2014

R2 v1 2026-06-22T03:12:03.508Z