Multiview Learning of Weighted Majority Vote by Bregman Divergence Minimization - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

Multiview Learning of Weighted Majority Vote by Bregman Divergence Minimization

Résumé

We tackle the issue of classifier combinations when observations have multiple views. Our method jointly learns view-specific weighted majority vote classifiers (i.e. for each view) over a set of base voters, and a second weighted majority vote classifier over the set of these view-specific weighted majority vote classifiers. We show that the empirical risk minimization of the final majority vote given a multiview training set can be cast as the minimization of Bregman divergences. This allows us to derive a parallel-update optimization algorithm for learning our multiview model. We empirically study our algorithm with a particular focus on the impact of the training set size on the multiview learning results. The experiments show that our approach is able to overcome the lack of labeled information.
Fichier principal
Vignette du fichier
main_report.pdf (921.18 Ko) Télécharger le fichier
MV_hierarchy.pdf (35.06 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01799173 , version 1 (24-05-2018)

Identifiants

Citer

Anil Goyal, Emilie Morvant, Massih-Reza Amini. Multiview Learning of Weighted Majority Vote by Bregman Divergence Minimization. International Symposium on Intelligent Data Analysis (IDA), Oct 2018, ‘s-Hertogenbosch, Netherlands. ⟨hal-01799173⟩
88 Consultations
142 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More