Margin error and generalization capabilities of multiclass discriminant systems

André Elisseeff 1 Yann Guermeur 2 Hélène Paugam-Moisy
2 CORTEX - Neuromimetic intelligence
INRIA Lorraine, LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : This technical report exposes an extension of a lemma by Bartlett to the multiclass case. First, discussing the notion of margin and how its use for biclass discriminant models, we introduce the notion of margin for the multiclass case. Then, we develop our calculations and prove how to derive generalization error bounds for classifiers having multiple outputs. These bounds are expressed in terms of covering numbers. This report shows how to bound these quantities for the set of functions we consider and give an example for linear discriminant systems.
Type de document :
Rapport
[Intern report] A00-R-112 || elisseeff00b, 2000
Liste complète des métadonnées

https://hal.inria.fr/inria-00099333
Contributeur : Publications Loria <>
Soumis le : mardi 26 septembre 2006 - 08:53:00
Dernière modification le : jeudi 11 janvier 2018 - 06:19:48

Identifiants

  • HAL Id : inria-00099333, version 1

Collections

Citation

André Elisseeff, Yann Guermeur, Hélène Paugam-Moisy. Margin error and generalization capabilities of multiclass discriminant systems. [Intern report] A00-R-112 || elisseeff00b, 2000. 〈inria-00099333〉

Partager

Métriques

Consultations de la notice

161