A PAC-Bayesian Approach for Domain Adaptation with Specialization to Linear Classifiers

Abstract : We provide a first PAC-Bayesian analysis for domain adaptation (DA) which arises when the learning and test distributions differ. It relies on a novel distribution pseudodistance based on a disagreement averaging. Using this measure, we derive a PAC-Bayesian DA bound for the stochastic Gibbs classifier. This bound has the advantage of being directly optimizable for any hypothesis space. We specialize it to linear classifiers, and design a learning algorithm which shows interesting results on a synthetic problem and on a popular sentiment annotation task. This opens the door to tackling DA tasks by making use of all the PAC-Bayesian tools.
Document type :
Conference papers
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-00822685
Contributor : Emilie Morvant <>
Submitted on : Thursday, May 16, 2013 - 2:49:57 PM
Last modification on : Tuesday, April 2, 2019 - 1:41:26 AM
Long-term archiving on : Saturday, August 17, 2013 - 2:50:11 AM

File

pbda.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00822685, version 1

Citation

Pascal Germain, Amaury Habrard, François Laviolette, Emilie Morvant. A PAC-Bayesian Approach for Domain Adaptation with Specialization to Linear Classifiers. International Conference on Machine Learning 2013, Jun 2013, Atlanta, United States. pp.738-746. ⟨hal-00822685⟩

Share

Metrics

Record views

567

Files downloads

1193