On the Generalization of the C-Bound to Structured Output Ensemble Methods - Archive ouverte HAL Accéder directement au contenu
Rapport (Rapport De Recherche) Année : 2015

On the Generalization of the C-Bound to Structured Output Ensemble Methods

Résumé

This paper generalizes an important result from the PAC-Bayesian literature for binary classification to the case of ensemble methods for structured outputs. We prove a generic version of the \Cbound, an upper bound over the risk of models expressed as a weighted majority vote that is based on the first and second statistical moments of the vote's margin. This bound may advantageously $(i)$ be applied on more complex outputs such as multiclass labels and multilabel, and $(ii)$ allow to consider margin relaxations. These results open the way to develop new ensemble methods for structured output prediction with PAC-Bayesian guarantees.
Fichier principal
Vignette du fichier
main_Cborne_multi.pdf (793.63 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-01054337 , version 1 (06-08-2014)
hal-01054337 , version 2 (15-06-2015)

Identifiants

Citer

François Laviolette, Emilie Morvant, Liva Ralaivola, Jean-Francis Roy. On the Generalization of the C-Bound to Structured Output Ensemble Methods. [Research Report] Laboratoire Hubert Curien, Université Jean Monnet Saint-Etienne; Laboratoire Informatique Fondamentale, Aix-Marseille Université; Département d'Informatique et de Génie Logiciel, Université Laval (Québec). 2015. ⟨hal-01054337v2⟩
419 Consultations
146 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More