A New PAC-Bayesian Perspective on Domain Adaptation

Pascal Germain 1, 2 Amaury Habrard 3 François Laviolette 2 Emilie Morvant 3
1 SIERRA - Statistical Machine Learning and Parsimony
DI-ENS - Département d'informatique de l'École normale supérieure, ENS Paris - École normale supérieure - Paris, Inria Paris-Rocquencourt, CNRS - Centre National de la Recherche Scientifique : UMR8548
Abstract : 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.
Document type :
Complete list of metadatas

Contributor : Emilie Morvant <>
Submitted on : Monday, March 14, 2016 - 5:01:11 PM
Last modification on : Thursday, February 7, 2019 - 2:42:37 PM


Files produced by the author(s)


  • HAL Id : hal-01163722, version 3
  • ARXIV : 1506.04573


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⟩



Record views


Files downloads