A Fast Method for Training Linear SVM in the Primal
Trinh Minh Tri Do and Thierry Artieres
European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases Volume 5211, pp. 272-287, 2008.

## Abstract

We propose a new algorithm for training a linear Support Vector Machine in the primal. The algorithm mixes ideas from non smooth optimization, subgradient methods, and cutting planes methods. This yields a fast algorithm that compares well to state of the art algorithms. It is proved to require $O(1/{\lambda\epsilon})$ iterations to converge to a solution with accuracy $\epsilon$. Additionally we provide an exact shrinking method in the primal that allows reducing the complexity of an iteration to much less than $O(N)$ where $N$ is the number of training samples.

EPrint Type: Article Project Keyword UNSPECIFIED Learning/Statistics & Optimisation 5059 Trinh Minh Tri Do 24 March 2009