Pegasos Algorithm for One-Class Support Vector Machine
Summary :
Training one-class support vector machines (one-class SVMs) involves solving a quadratic programming (QP) problem. By increasing the number of training samples, solving this QP problem becomes intractable. In this paper, we describe a modified Pegasos algorithm for fast training of one-class SVMs. We show that this algorithm is much faster than the standard one-class SVM without loss of performance in the case of linear kernel.
- Publication
- IEICE TRANSACTIONS on Information Vol.E96-D No.5 pp.1223-1226
- Publication Date
- 2013/05/01
- Publicized
- Online ISSN
- 1745-1361
- DOI
- 10.1587/transinf.E96.D.1223
- Type of Manuscript
- LETTER
- Category
- Artificial Intelligence, Data Mining
Authors
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Changki LEE, "Pegasos Algorithm for One-Class Support Vector Machine" in IEICE TRANSACTIONS on Information,
vol. E96-D, no. 5, pp. 1223-1226, May 2013, doi: 10.1587/transinf.E96.D.1223.
Abstract: Training one-class support vector machines (one-class SVMs) involves solving a quadratic programming (QP) problem. By increasing the number of training samples, solving this QP problem becomes intractable. In this paper, we describe a modified Pegasos algorithm for fast training of one-class SVMs. We show that this algorithm is much faster than the standard one-class SVM without loss of performance in the case of linear kernel.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E96.D.1223/_p
Copy
@ARTICLE{e96-d_5_1223,
author={Changki LEE, },
journal={IEICE TRANSACTIONS on Information},
title={Pegasos Algorithm for One-Class Support Vector Machine},
year={2013},
volume={E96-D},
number={5},
pages={1223-1226},
abstract={Training one-class support vector machines (one-class SVMs) involves solving a quadratic programming (QP) problem. By increasing the number of training samples, solving this QP problem becomes intractable. In this paper, we describe a modified Pegasos algorithm for fast training of one-class SVMs. We show that this algorithm is much faster than the standard one-class SVM without loss of performance in the case of linear kernel.},
keywords={},
doi={10.1587/transinf.E96.D.1223},
ISSN={1745-1361},
month={May},}
Copy
TY - JOUR
TI - Pegasos Algorithm for One-Class Support Vector Machine
T2 - IEICE TRANSACTIONS on Information
SP - 1223
EP - 1226
AU - Changki LEE
PY - 2013
DO - 10.1587/transinf.E96.D.1223
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E96-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2013
AB - Training one-class support vector machines (one-class SVMs) involves solving a quadratic programming (QP) problem. By increasing the number of training samples, solving this QP problem becomes intractable. In this paper, we describe a modified Pegasos algorithm for fast training of one-class SVMs. We show that this algorithm is much faster than the standard one-class SVM without loss of performance in the case of linear kernel.
ER -
English
