We find very tight bounds on the accuracy of a Support Vector Machine classification error within the Algorithmic Inference framework. The framework is specially suitable for this kind of classifier since (i) we know the number of support vectors really employed, as an ancillary output of the learning procedure, and (ii) we can appreciate confidence intervals of misclassifying probability exactly in function of the cardinality of these vectors. As a result we obtain confidence intervals that are up to an order narrower than those supplied in the literature, having a slight different meaning due to the different approach they come from, but the same operational function. We numerically check the covering of these intervals.

Tight Bounds for SVM Classification Error / B. Apolloni, S. Bassis, S. Gaito, D. Malchiodi - In: Proceedings of the 2005 International Conference on Neural Networks & Brain : October 13-15, 2005, Beijing, China / Mingsheng Zhao and Zhongzhi Shi. - Piscataway, NJ : IEEE Computer Society, 2005. - ISBN 0780394224. - pp. 5-8 (( convegno International conference on Neural Networks & Brain tenutosi a Beijing, China nel 2005.

Tight Bounds for SVM Classification Error

B. Apolloni
Primo
;
S. Bassis
Secondo
;
S. Gaito
Penultimo
;
D. Malchiodi
Ultimo
2005

Abstract

We find very tight bounds on the accuracy of a Support Vector Machine classification error within the Algorithmic Inference framework. The framework is specially suitable for this kind of classifier since (i) we know the number of support vectors really employed, as an ancillary output of the learning procedure, and (ii) we can appreciate confidence intervals of misclassifying probability exactly in function of the cardinality of these vectors. As a result we obtain confidence intervals that are up to an order narrower than those supplied in the literature, having a slight different meaning due to the different approach they come from, but the same operational function. We numerically check the covering of these intervals.
Settore INF/01 - Informatica
IEEE
China Neural Networks Council
Book Part (author)
File in questo prodotto:
File Dimensione Formato  
IEEE-Published.pdf

accesso aperto

Tipologia: Altro
Dimensione 2.01 MB
Formato Adobe PDF
2.01 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

Caricamento pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/2434/7854
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? ND
social impact