Multiview Boosting by Controlling the Diversity and the Accuracy of View-specific Voters
Abstract
In this paper we propose a boosting based multiview learning algorithm, referred to as PB-MVBoost, which iteratively learns i) weights over view-specific voters capturing view-specific information; and ii) weights over views by optimizing a PAC-Bayes multiview C-Bound that takes into account the accuracy of view-specific classifiers and the diversity between the views. We derive a generalization bound for this strategy following the PAC-Bayes theory which is a suitable tool to deal with models expressed as weighted combination over a set of voters. Different experiments on three publicly available datasets show the efficiency of the proposed approach with respect to state-of-art models.
Domains
Machine Learning [stat.ML]
Fichier principal
PBMVBoost_Paper.pdf (1.29 Mo)
Télécharger le fichier
MNIST_1_500_iterations.png (68.5 Ko)
Télécharger le fichier
MNIST_1_F1.png (137.24 Ko)
Télécharger le fichier
MNIST_1_acc.png (117.92 Ko)
Télécharger le fichier
MNIST_2_500_iterations.png (63.7 Ko)
Télécharger le fichier
MNIST_2_F1.png (137.38 Ko)
Télécharger le fichier
MNIST_2_acc.png (119.83 Ko)
Télécharger le fichier
Reuters_500_iterations.png (70.67 Ko)
Télécharger le fichier
Reuters_F1.png (142.82 Ko)
Télécharger le fichier
Reuters_acc.png (119.62 Ko)
Télécharger le fichier
hierarchy.pdf (19.34 Ko)
Télécharger le fichier
histograms.png (73.37 Ko)
Télécharger le fichier
Origin | Files produced by the author(s) |
---|
Origin | Files produced by the author(s) |
---|
Origin | Files produced by the author(s) |
---|