Global convergence analysis of decomposition methods for support vector regression

Jun Guo; Norikazu Takahashi

doi:10.1007/978-3-540-87732-5_74

Global convergence analysis of decomposition methods for support vector regression

Jun Guo, Norikazu Takahashi

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

1 Citation (Scopus)

Abstract

Decomposition method has been widely used to efficiently solve the large size quadratic programming (QP) problems arising in support vector regression (SVR). In a decomposition method, a large QP problem is decomposed into a series of smaller QP subproblems, which can be solved much faster than the original one. In this paper, we analyze the global convergence of decomposition methods for SVR. We will show the decomposition methods for the convex programming problem formulated by Flake and Lawrence always stop within a finite number of iterations.

Original language	English
Title of host publication	Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings
Publisher	Springer Verlag
Pages	663-673
Number of pages	11
Edition	PART 1
ISBN (Print)	3540877312, 9783540877318
DOIs	https://doi.org/10.1007/978-3-540-87732-5_74
Publication status	Published - 2008
Externally published	Yes
Event	5th International Symposium on Neural Networks, ISNN 2008 - Beijing, China Duration: Sept 24 2008 → Sept 28 2008

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 1
Volume	5263 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	5th International Symposium on Neural Networks, ISNN 2008
Country/Territory	China
City	Beijing
Period	9/24/08 → 9/28/08

Keywords

Decomposition method
Global convergence
Support vector regression

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/978-3-540-87732-5_74

Cite this

Guo, J., & Takahashi, N. (2008). Global convergence analysis of decomposition methods for support vector regression. In Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings (PART 1 ed., pp. 663-673). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5263 LNCS, No. PART 1). Springer Verlag. https://doi.org/10.1007/978-3-540-87732-5_74

Global convergence analysis of decomposition methods for support vector regression. / Guo, Jun; Takahashi, Norikazu.
Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings. PART 1. ed. Springer Verlag, 2008. p. 663-673 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5263 LNCS, No. PART 1).

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

Guo, J & Takahashi, N 2008, Global convergence analysis of decomposition methods for support vector regression. in Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings. PART 1 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 1, vol. 5263 LNCS, Springer Verlag, pp. 663-673, 5th International Symposium on Neural Networks, ISNN 2008, Beijing, China, 9/24/08. https://doi.org/10.1007/978-3-540-87732-5_74

Guo J, Takahashi N. Global convergence analysis of decomposition methods for support vector regression. In Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings. PART 1 ed. Springer Verlag. 2008. p. 663-673. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1). doi: 10.1007/978-3-540-87732-5_74

Guo, Jun ; Takahashi, Norikazu. / Global convergence analysis of decomposition methods for support vector regression. Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings. PART 1. ed. Springer Verlag, 2008. pp. 663-673 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1).

@inproceedings{7852cb5fd6b14efb9d35274a82d1abaf,

title = "Global convergence analysis of decomposition methods for support vector regression",

abstract = "Decomposition method has been widely used to efficiently solve the large size quadratic programming (QP) problems arising in support vector regression (SVR). In a decomposition method, a large QP problem is decomposed into a series of smaller QP subproblems, which can be solved much faster than the original one. In this paper, we analyze the global convergence of decomposition methods for SVR. We will show the decomposition methods for the convex programming problem formulated by Flake and Lawrence always stop within a finite number of iterations.",

keywords = "Decomposition method, Global convergence, Support vector regression",

author = "Jun Guo and Norikazu Takahashi",

year = "2008",

doi = "10.1007/978-3-540-87732-5_74",

language = "English",

isbn = "3540877312",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

number = "PART 1",

pages = "663--673",

booktitle = "Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings",

edition = "PART 1",

}

TY - GEN

T1 - Global convergence analysis of decomposition methods for support vector regression

AU - Guo, Jun

AU - Takahashi, Norikazu

PY - 2008

Y1 - 2008

N2 - Decomposition method has been widely used to efficiently solve the large size quadratic programming (QP) problems arising in support vector regression (SVR). In a decomposition method, a large QP problem is decomposed into a series of smaller QP subproblems, which can be solved much faster than the original one. In this paper, we analyze the global convergence of decomposition methods for SVR. We will show the decomposition methods for the convex programming problem formulated by Flake and Lawrence always stop within a finite number of iterations.

AB - Decomposition method has been widely used to efficiently solve the large size quadratic programming (QP) problems arising in support vector regression (SVR). In a decomposition method, a large QP problem is decomposed into a series of smaller QP subproblems, which can be solved much faster than the original one. In this paper, we analyze the global convergence of decomposition methods for SVR. We will show the decomposition methods for the convex programming problem formulated by Flake and Lawrence always stop within a finite number of iterations.

KW - Decomposition method

KW - Global convergence

KW - Support vector regression

UR - http://www.scopus.com/inward/record.url?scp=59149088377&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=59149088377&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-87732-5_74

DO - 10.1007/978-3-540-87732-5_74

M3 - Conference contribution

AN - SCOPUS:59149088377

SN - 3540877312

SN - 9783540877318

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 663

EP - 673

BT - Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings

PB - Springer Verlag

T2 - 5th International Symposium on Neural Networks, ISNN 2008

Y2 - 24 September 2008 through 28 September 2008

ER -

Global convergence analysis of decomposition methods for support vector regression

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this