Support vectors algorithms as regularization networks

Andrea Caponnetto; Lorenzo Rosasco; Francesca Odone; Alessandro Verri

Support vectors algorithms as regularization networks

Andrea Caponnetto, Lorenzo Rosasco, Francesca Odone, Alessandro Verri

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

1 Citation (Scopus)

Abstract

In this paper we show that several Support Vector methods, including one-class SVM and a number of non-standard SVM classification techniques, can be viewed as special implementations of a general regularization network. Formally, the connection is obtained by choosing the appropriate loss function and parametrized by the exponent of the offset in the penalty term. The mathematical properties of the underlying algorithms can then be more conveniently studied within the theoretical framework of regularization networks.

Original language	English
Title of host publication	ESANN 2005 Proceedings - 13th European Symposium on Artificial Neural Networks
Pages	595-600
Publication status	Published - 2007
Externally published	Yes
Event	13th European Symposium on Artificial Neural Networks, ESANN 2005 - Bruges, Belgium Duration: 27 Apr 2005 → 29 Apr 2005

Conference

Conference	13th European Symposium on Artificial Neural Networks, ESANN 2005
Place	Belgium
City	Bruges
Period	27/04/05 → 29/04/05

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{3d7e95c86e974b208098cc968f221121,

title = "Support vectors algorithms as regularization networks",

abstract = "In this paper we show that several Support Vector methods, including one-class SVM and a number of non-standard SVM classification techniques, can be viewed as special implementations of a general regularization network. Formally, the connection is obtained by choosing the appropriate loss function and parametrized by the exponent of the offset in the penalty term. The mathematical properties of the underlying algorithms can then be more conveniently studied within the theoretical framework of regularization networks.",

author = "Andrea Caponnetto and Lorenzo Rosasco and Francesca Odone and Alessandro Verri",

year = "2007",

language = "English",

isbn = "2930307056",

pages = "595--600",

booktitle = "ESANN 2005 Proceedings - 13th European Symposium on Artificial Neural Networks",

note = "13th European Symposium on Artificial Neural Networks, ESANN 2005 ; Conference date: 27-04-2005 Through 29-04-2005",

}

TY - GEN

T1 - Support vectors algorithms as regularization networks

AU - Caponnetto, Andrea

AU - Rosasco, Lorenzo

AU - Odone, Francesca

AU - Verri, Alessandro

PY - 2007

Y1 - 2007

N2 - In this paper we show that several Support Vector methods, including one-class SVM and a number of non-standard SVM classification techniques, can be viewed as special implementations of a general regularization network. Formally, the connection is obtained by choosing the appropriate loss function and parametrized by the exponent of the offset in the penalty term. The mathematical properties of the underlying algorithms can then be more conveniently studied within the theoretical framework of regularization networks.

AB - In this paper we show that several Support Vector methods, including one-class SVM and a number of non-standard SVM classification techniques, can be viewed as special implementations of a general regularization network. Formally, the connection is obtained by choosing the appropriate loss function and parametrized by the exponent of the offset in the penalty term. The mathematical properties of the underlying algorithms can then be more conveniently studied within the theoretical framework of regularization networks.

UR - http://www.scopus.com/inward/record.url?scp=84887008319&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-84887008319&origin=recordpage

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 2930307056

SN - 9782930307053

SP - 595

EP - 600

BT - ESANN 2005 Proceedings - 13th European Symposium on Artificial Neural Networks

T2 - 13th European Symposium on Artificial Neural Networks, ESANN 2005

Y2 - 27 April 2005 through 29 April 2005

ER -

Support vectors algorithms as regularization networks

Abstract

Conference

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this