A Simple Decomposition Method for Support Vector Machines
Resource
Machine Learning 46 (1-3): 291-314
Journal
Machine Learning
Journal Issue
46
Pages
291-314
Date Issued
2002
Date
2002
Author(s)
Hsu, Chih-Wei
Abstract
The decomposition method is currently one of the major methods for solving support vector machines. An important issue of this method is the selection of working sets. In this paper through the design of decomposition methods for bound-constrained SVM formulations we demonstrate that the working set selection is not a trivial task. Then from the experimental analysis we propose a simple selection of the working set which leads to faster convergences for difficult cases. Numerical experiments on different types of problems are conducted to demonstrate the viability of the proposed method.
Type
journal article
File(s)![Thumbnail Image]()
Loading...
Name
11.pdf
Size
178.1 KB
Format
Adobe PDF
Checksum
(MD5):a35cacb6681fc9c8a817c2f381ef8488