PAC-Bayes and Domain Adaptation - Université de Lille Accéder directement au contenu
Article Dans Une Revue Neurocomputing Année : 2020

PAC-Bayes and Domain Adaptation

Résumé

We provide two main contributions in PAC-Bayesian theory for domain adaptation where the objective is to learn, from a source distribution, a well-performing majority vote on a different, but related, target distribution. Firstly, we propose an improvement of the previous approach we proposed in Germain et al. (2013), which relies on a novel distribution pseudodistance based on a disagreement averaging, allowing us to derive a new tighter domain adaptation bound for the target risk. While this bound stands in the spirit of common domain adaptation works, we derive a second bound (introduced in Germain et al., 2016) that brings a new perspective on domain adaptation 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. We discuss and compare both results, from which we obtain PAC-Bayesian generalization bounds. Furthermore, from the PAC-Bayesian specialization to linear classifiers, we infer two learning algorithms, and we evaluate them on real data.
Fichier principal
Vignette du fichier
main.pdf (1.55 Mo) Télécharger le fichier
Loading...

Dates et versions

hal-01563152 , version 1 (17-07-2017)
hal-01563152 , version 2 (06-11-2018)
hal-01563152 , version 3 (06-11-2019)

Identifiants

Citer

Pascal Germain, Amaury Habrard, François Laviolette, Emilie Morvant. PAC-Bayes and Domain Adaptation. Neurocomputing, 2020, 379, pp.379-397. ⟨10.1016/j.neucom.2019.10.105⟩. ⟨hal-01563152v3⟩
325 Consultations
519 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More