Method of parallel sequential minimal optimization for fast training support vector machine

Liyan Tian; Xiaoguang Hu

doi:10.4028/www.scientific.net/AMM.29-32.947

Method of parallel sequential minimal optimization for fast training support vector machine

Liyan Tian^*
, Xiaoguang Hu

^*Corresponding author for this work

School of Automation Science and Electrical Engineering

Beihang University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Scopus citations

Abstract

A fast training support vector machine using parallel sequential minimal optimization is presented in this paper. Up to now, sequential minimal optimization (SMO) is one of the major algorithms for training SVM, but it still requires a large amount of computation time for the large sample problems. Unlike the traditional SMO, the parallel SMO partitions the entire training data set into small subsets first and then runs multiple CPU processors to seal with each of the partitioned data set. Experiments show that the new algorithm has great advantage in terms of speediness when applied to problems with large training sets and high dimensional spaces without reducing generalization performance of SVM.

Original language	English
Title of host publication	Applied Mechanics and Mechanical Engineering
Pages	947-951
Number of pages	5
DOIs	https://doi.org/10.4028/www.scientific.net/AMM.29-32.947
State	Published - 2010
Event	2010 International Conference on Applied Mechanics and Mechanical Engineering, ICAMME 2010 - Changsha, China Duration: 8 Sep 2010 → 9 Sep 2010

Publication series

Name	Applied Mechanics and Materials
Volume	29-32
ISSN (Print)	1660-9336
ISSN (Electronic)	1662-7482

Conference

Conference	2010 International Conference on Applied Mechanics and Mechanical Engineering, ICAMME 2010
Country/Territory	China
City	Changsha
Period	8/09/10 → 9/09/10

Keywords

Parallel algorithm
Sequential minimal optimization(SMO)
Support vector machine(SVM)

Access to Document

10.4028/www.scientific.net/AMM.29-32.947

Cite this

@inproceedings{d2eaafec794b4078b30d497d26ee0d6f,

title = "Method of parallel sequential minimal optimization for fast training support vector machine",

abstract = "A fast training support vector machine using parallel sequential minimal optimization is presented in this paper. Up to now, sequential minimal optimization (SMO) is one of the major algorithms for training SVM, but it still requires a large amount of computation time for the large sample problems. Unlike the traditional SMO, the parallel SMO partitions the entire training data set into small subsets first and then runs multiple CPU processors to seal with each of the partitioned data set. Experiments show that the new algorithm has great advantage in terms of speediness when applied to problems with large training sets and high dimensional spaces without reducing generalization performance of SVM.",

keywords = "Parallel algorithm, Sequential minimal optimization(SMO), Support vector machine(SVM)",

author = "Liyan Tian and Xiaoguang Hu",

year = "2010",

doi = "10.4028/www.scientific.net/AMM.29-32.947",

language = "英语",

isbn = "9780878492459",

series = "Applied Mechanics and Materials",

pages = "947--951",

booktitle = "Applied Mechanics and Mechanical Engineering",

note = "2010 International Conference on Applied Mechanics and Mechanical Engineering, ICAMME 2010 ; Conference date: 08-09-2010 Through 09-09-2010",

}

Tian, L & Hu, X 2010, Method of parallel sequential minimal optimization for fast training support vector machine. in Applied Mechanics and Mechanical Engineering. Applied Mechanics and Materials, vol. 29-32, pp. 947-951, 2010 International Conference on Applied Mechanics and Mechanical Engineering, ICAMME 2010, Changsha, China, 8/09/10. https://doi.org/10.4028/www.scientific.net/AMM.29-32.947

TY - GEN

T1 - Method of parallel sequential minimal optimization for fast training support vector machine

AU - Tian, Liyan

AU - Hu, Xiaoguang

PY - 2010

Y1 - 2010

N2 - A fast training support vector machine using parallel sequential minimal optimization is presented in this paper. Up to now, sequential minimal optimization (SMO) is one of the major algorithms for training SVM, but it still requires a large amount of computation time for the large sample problems. Unlike the traditional SMO, the parallel SMO partitions the entire training data set into small subsets first and then runs multiple CPU processors to seal with each of the partitioned data set. Experiments show that the new algorithm has great advantage in terms of speediness when applied to problems with large training sets and high dimensional spaces without reducing generalization performance of SVM.

AB - A fast training support vector machine using parallel sequential minimal optimization is presented in this paper. Up to now, sequential minimal optimization (SMO) is one of the major algorithms for training SVM, but it still requires a large amount of computation time for the large sample problems. Unlike the traditional SMO, the parallel SMO partitions the entire training data set into small subsets first and then runs multiple CPU processors to seal with each of the partitioned data set. Experiments show that the new algorithm has great advantage in terms of speediness when applied to problems with large training sets and high dimensional spaces without reducing generalization performance of SVM.

KW - Parallel algorithm

KW - Sequential minimal optimization(SMO)

KW - Support vector machine(SVM)

UR - https://www.scopus.com/pages/publications/78650795038

U2 - 10.4028/www.scientific.net/AMM.29-32.947

DO - 10.4028/www.scientific.net/AMM.29-32.947

M3 - 会议稿件

AN - SCOPUS:78650795038

SN - 9780878492459

T3 - Applied Mechanics and Materials

SP - 947

EP - 951

BT - Applied Mechanics and Mechanical Engineering

T2 - 2010 International Conference on Applied Mechanics and Mechanical Engineering, ICAMME 2010

Y2 - 8 September 2010 through 9 September 2010

ER -

Method of parallel sequential minimal optimization for fast training support vector machine

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this