A New PAC-Bayesian Perspective on Domain Adaptation - Archive ouverte HAL Accéder directement au contenu
Rapport (Rapport De Recherche) Année : 2015

A New PAC-Bayesian Perspective on Domain Adaptation

Résumé

We study the issue of PAC-Bayesian domain adaptation: We want to learn, from a source domain, a majority vote model dedicated to a target one. Our theoretical contribution brings a new perspective by deriving an upper-bound on the target risk where the distributions' divergence---expressed as a ratio---controls the trade-off between a source error measure and the target voters' disagreement. Our bound suggests that one has to focus on regions where the source data is informative. From this result, we derive a PAC-Bayesian generalization bound, and specialize it to linear classifiers. Then, we infer a learning algorithm and perform experiments on real data.
Fichier principal
Vignette du fichier
DALC_report_V3.pdf (603.07 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-01163722 , version 1 (15-06-2015)
hal-01163722 , version 2 (21-09-2015)
hal-01163722 , version 3 (14-03-2016)

Identifiants

Citer

Pascal Germain, Amaury Habrard, François Laviolette, Emilie Morvant. A New PAC-Bayesian Perspective on Domain Adaptation. [Research Report] Univ Lyon, UJM-Saint-Etienne, CNRS, Laboratoire Hubert Curien UMR 5516, F-42023 Saint-Etienne, France; Département d'informatique et de génie logiciel, Université Laval (Québec); INRIA - Sierra Project-Team, Ecole Normale Sup´erieure, Paris, France. 2015. ⟨hal-01163722v3⟩
196 Consultations
105 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More