Fisher Consistency of Multicategory Support Vector Machines

Yufeng Liu
Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics, PMLR 2:291-298, 2007.

Abstract

The Support Vector Machine (SVM) has become one of the most popular machine learning techniques in recent years. The success of the SVM is mostly due to its elegant margin concept and theory in binary classification. Generalization to the multicategory setting, however, is not trivial. There are a number of different multicategory extensions of the SVM in the literature. In this paper, we review several commonly used extensions and Fisher consistency of these extensions. For inconsistent extensions, we propose two approaches to make them Fisher consistent, one is to add bounded constraints and the other is to truncate unbounded hinge losses.

Cite this Paper

BibTeX
@InProceedings{pmlr-v2-liu07b, title = {Fisher Consistency of Multicategory Support Vector Machines}, author = {Liu, Yufeng}, booktitle = {Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics}, pages = {291--298}, year = {2007}, editor = {Meila, Marina and Shen, Xiaotong}, volume = {2}, series = {Proceedings of Machine Learning Research}, address = {San Juan, Puerto Rico}, month = {21--24 Mar}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v2/liu07b/liu07b.pdf}, url = {https://proceedings.mlr.press/v2/liu07b.html}, abstract = {The Support Vector Machine (SVM) has become one of the most popular machine learning techniques in recent years. The success of the SVM is mostly due to its elegant margin concept and theory in binary classification. Generalization to the multicategory setting, however, is not trivial. There are a number of different multicategory extensions of the SVM in the literature. In this paper, we review several commonly used extensions and Fisher consistency of these extensions. For inconsistent extensions, we propose two approaches to make them Fisher consistent, one is to add bounded constraints and the other is to truncate unbounded hinge losses.} }
Endnote
%0 Conference Paper %T Fisher Consistency of Multicategory Support Vector Machines %A Yufeng Liu %B Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2007 %E Marina Meila %E Xiaotong Shen %F pmlr-v2-liu07b %I PMLR %P 291--298 %U https://proceedings.mlr.press/v2/liu07b.html %V 2 %X The Support Vector Machine (SVM) has become one of the most popular machine learning techniques in recent years. The success of the SVM is mostly due to its elegant margin concept and theory in binary classification. Generalization to the multicategory setting, however, is not trivial. There are a number of different multicategory extensions of the SVM in the literature. In this paper, we review several commonly used extensions and Fisher consistency of these extensions. For inconsistent extensions, we propose two approaches to make them Fisher consistent, one is to add bounded constraints and the other is to truncate unbounded hinge losses.
RIS
TY - CPAPER TI - Fisher Consistency of Multicategory Support Vector Machines AU - Yufeng Liu BT - Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics DA - 2007/03/11 ED - Marina Meila ED - Xiaotong Shen ID - pmlr-v2-liu07b PB - PMLR DP - Proceedings of Machine Learning Research VL - 2 SP - 291 EP - 298 L1 - http://proceedings.mlr.press/v2/liu07b/liu07b.pdf UR - https://proceedings.mlr.press/v2/liu07b.html AB - The Support Vector Machine (SVM) has become one of the most popular machine learning techniques in recent years. The success of the SVM is mostly due to its elegant margin concept and theory in binary classification. Generalization to the multicategory setting, however, is not trivial. There are a number of different multicategory extensions of the SVM in the literature. In this paper, we review several commonly used extensions and Fisher consistency of these extensions. For inconsistent extensions, we propose two approaches to make them Fisher consistent, one is to add bounded constraints and the other is to truncate unbounded hinge losses. ER -
APA
Liu, Y.. (2007). Fisher Consistency of Multicategory Support Vector Machines. Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 2:291-298 Available from https://proceedings.mlr.press/v2/liu07b.html.

Related Material