Skip to main navigation Skip to search Skip to main content

A fast training algorithm for least squares SVM

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

A fast training algorithm for Least Squares SVM (LS-SVM) classifiers was proposed, which is based on incremental and decremental learning theory. When a SV (Support Vector) is added or removed, computation based on previous training result replaces large-scale matrix inverse, thus the computation cost is reduced. The innovation is that by reasonable use of incremental and decremental learning the proposed algorithm can adaptively adjust the size of training sets (number of SVs) according to the specific classification problem. Finally several experiments show the validity of proposed algorithm.

Original languageEnglish
Title of host publicationProceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.
Pages586-589
Number of pages4
DOIs
StatePublished - 2007
Event3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007 - Kaohsiung, Taiwan, Province of China
Duration: 26 Nov 200728 Nov 2007

Publication series

NameProceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.
Volume2

Conference

Conference3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007
Country/TerritoryTaiwan, Province of China
CityKaohsiung
Period26/11/0728/11/07

Fingerprint

Dive into the research topics of 'A fast training algorithm for least squares SVM'. Together they form a unique fingerprint.

Cite this