Rademacher Complexity and Generalization Performance of Multi-category Margin Classifiers - Department of Algorithms, Computation, Image and Geometry Accéder directement au contenu
Article Dans Une Revue Neurocomputing Année : 2019

Rademacher Complexity and Generalization Performance of Multi-category Margin Classifiers

Khadija Musayeva
  • Fonction : Auteur
  • PersonId : 1039849
Fabien Lauer
Yann Guermeur
  • Fonction : Auteur
  • PersonId : 830806

Résumé

One of the main open problems in the theory of multi-category margin classification is the form of the optimal dependency of a guaranteed risk on the number C of categories, the sample size m and the margin parameter gamma. From a practical point of view, the theoretical analysis of generalization performance contributes to the development of new learning algorithms. In this paper, we focus only on the theoretical aspect of the question posed. More precisely, under minimal learnability assumptions, we derive a new risk bound for multi-category margin classifiers. We improve the dependency on C over the state of the art when the margin loss function considered satisfies the Lipschitz condition. We start with the basic supremum inequality that involves a Rademacher complexity as a capacity measure. This capacity measure is then linked to the metric entropy through the chaining method. In this context, our improvement is based on the introduction of a new combinatorial metric entropy bound.
Fichier principal
Vignette du fichier
Musayeva18.pdf (267.62 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-01941089 , version 1 (30-11-2018)

Identifiants

Citer

Khadija Musayeva, Fabien Lauer, Yann Guermeur. Rademacher Complexity and Generalization Performance of Multi-category Margin Classifiers. Neurocomputing, 2019, 342, pp.6-15. ⟨10.1016/j.neucom.2018.11.096⟩. ⟨hal-01941089⟩
78 Consultations
1164 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More