Personal tools
You are here: Home Publications On the convergence of the decomposition methods for support vector machines
Document Actions

Chih-Jen Lin (2000)

On the convergence of the decomposition methods for support vector machines

NTU.

The decomposition method is currently one of the major methods for solving support vector machines (SVM). Its convergence properties have not been fully understood. The general asymptotic convergence was first proposed by Chang et al. However, their working set selection does not coincide with existing implementation. A later breakthrough by Keerthi and Gilbert proved the convergence for practical cases while the size of the working set is restricted to two. In this paper, we prove the convergence of the algorithm used by the software SVMlight and other later implementation such as mySVM and SVMTorch. The size of the working set can be any even number. Extensions to other SVM formulations are also discussed.

by admin last modified 2007-01-31 11:08

Powered by Plone CMS, the Open Source Content Management System