This paper points out an important source of confusion and inefficiency in Smola and
Scholkopf's Sequential Minimal Optimization (SMO) algorithm for SVM regression that is
caused by the use of a single threshold value. Using clues from the KKT conditions for the
dual problem, two threshold parameters are employed to derive modifications of SMO for regression.
These modified algorithms perform significantly faster than the original SMO on the
datasets tried.
1 Introduction
Support Vector...
%0 Journal Article
%1 citeulike:678623
%A Shevade, S. K.
%A Keerthi, S. S.
%A Bhattacharyya, C.
%A Murthy, K. R. K.
%D 2000
%J IEEE-NN
%K algorithm smo
%N 5
%T Improvements to the SMO Algorithm for SVM Regression
%U http://citeseer.ist.psu.edu/434628.html
%V 11
%X This paper points out an important source of confusion and inefficiency in Smola and
Scholkopf's Sequential Minimal Optimization (SMO) algorithm for SVM regression that is
caused by the use of a single threshold value. Using clues from the KKT conditions for the
dual problem, two threshold parameters are employed to derive modifications of SMO for regression.
These modified algorithms perform significantly faster than the original SMO on the
datasets tried.
1 Introduction
Support Vector...
@article{citeulike:678623,
abstract = {This paper points out an important source of confusion and inefficiency in Smola and
Scholkopf's Sequential Minimal Optimization (SMO) algorithm for SVM regression that is
caused by the use of a single threshold value. Using clues from the KKT conditions for the
dual problem, two threshold parameters are employed to derive modifications of SMO for regression.
These modified algorithms perform significantly faster than the original SMO on the
datasets tried.
1 Introduction
Support Vector...},
added-at = {2007-08-18T13:22:24.000+0200},
author = {Shevade, S. K. and Keerthi, S. S. and Bhattacharyya, C. and Murthy, K. R. K.},
biburl = {https://www.bibsonomy.org/bibtex/27687a7bf9af701d82e886a2e72de4ccf/a_olympia},
citeulike-article-id = {678623},
description = {citeulike},
interhash = {d54943fbcf9eb31a28ba3bc00b6603e6},
intrahash = {7687a7bf9af701d82e886a2e72de4ccf},
journal = {IEEE-NN},
keywords = {algorithm smo},
month = {September},
number = 5,
priority = {2},
timestamp = {2007-08-18T13:22:35.000+0200},
title = {Improvements to the SMO Algorithm for {SVM} Regression},
url = {http://citeseer.ist.psu.edu/434628.html},
volume = 11,
year = 2000
}