Rigorous proof of termination of SMO algorithm for support vector machines

Norikazu Takahashi; Tetsuo Nishi

doi:10.1109/TNN.2005.844857

Rigorous proof of termination of SMO algorithm for support vector machines

Norikazu Takahashi, Tetsuo Nishi

Research output: Contribution to journal › Article

35 Citations (Scopus)

Abstract

Sequential minimal optimization (SMO) algorithm is one of the simplest decomposition methods for learning of support vector machines (SVMs). Keerthi and Gilbert have recently studied the convergence property of SMO algorithm and given a proof that SMO algorithm always stops within a finite number of iterations. In this letter, we point out the incompleteness of their proof and give a more rigorous proof.

Original language	English
Pages (from-to)	774-776
Number of pages	3
Journal	IEEE Transactions on Neural Networks
Volume	16
Issue number	3
DOIs	https://doi.org/10.1109/TNN.2005.844857
Publication status	Published - May 2005
Externally published	Yes

Fingerprint

Keywords

Convergence
Sequential minimal optimization (SMO) algorithm
Support vector machines (SVMs)
Termination

ASJC Scopus subject areas

Control and Systems Engineering
Theoretical Computer Science
Electrical and Electronic Engineering
Artificial Intelligence
Computational Theory and Mathematics
Hardware and Architecture

Cite this

Takahashi, N., & Nishi, T. (2005). Rigorous proof of termination of SMO algorithm for support vector machines. IEEE Transactions on Neural Networks, 16(3), 774-776. https://doi.org/10.1109/TNN.2005.844857

@article{6434c982deb140ed8cdd3b6925e4cab2,

title = "Rigorous proof of termination of SMO algorithm for support vector machines",

abstract = "Sequential minimal optimization (SMO) algorithm is one of the simplest decomposition methods for learning of support vector machines (SVMs). Keerthi and Gilbert have recently studied the convergence property of SMO algorithm and given a proof that SMO algorithm always stops within a finite number of iterations. In this letter, we point out the incompleteness of their proof and give a more rigorous proof.",

keywords = "Convergence, Sequential minimal optimization (SMO) algorithm, Support vector machines (SVMs), Termination",

author = "Norikazu Takahashi and Tetsuo Nishi",

year = "2005",

month = may

doi = "10.1109/TNN.2005.844857",

language = "English",

volume = "16",

pages = "774--776",

journal = "IEEE Transactions on Neural Networks and Learning Systems",

issn = "2162-237X",

publisher = "IEEE Computational Intelligence Society",

number = "3",

}

TY - JOUR

T1 - Rigorous proof of termination of SMO algorithm for support vector machines

AU - Takahashi, Norikazu

AU - Nishi, Tetsuo

PY - 2005/5

Y1 - 2005/5

N2 - Sequential minimal optimization (SMO) algorithm is one of the simplest decomposition methods for learning of support vector machines (SVMs). Keerthi and Gilbert have recently studied the convergence property of SMO algorithm and given a proof that SMO algorithm always stops within a finite number of iterations. In this letter, we point out the incompleteness of their proof and give a more rigorous proof.

AB - Sequential minimal optimization (SMO) algorithm is one of the simplest decomposition methods for learning of support vector machines (SVMs). Keerthi and Gilbert have recently studied the convergence property of SMO algorithm and given a proof that SMO algorithm always stops within a finite number of iterations. In this letter, we point out the incompleteness of their proof and give a more rigorous proof.

KW - Convergence

KW - Sequential minimal optimization (SMO) algorithm

KW - Support vector machines (SVMs)

KW - Termination

UR - http://www.scopus.com/inward/record.url?scp=19344375172&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=19344375172&partnerID=8YFLogxK

U2 - 10.1109/TNN.2005.844857

DO - 10.1109/TNN.2005.844857

M3 - Article

C2 - 15941003

AN - SCOPUS:19344375172

VL - 16

SP - 774

EP - 776

JO - IEEE Transactions on Neural Networks and Learning Systems

JF - IEEE Transactions on Neural Networks and Learning Systems

SN - 2162-237X

IS - 3

ER -

Access to Document

10.1109/TNN.2005.844857