PAC-Bayesian Theorems for Domain Adaptation with Specialization to Linear Classifiers

Pascal Germain (1) , Amaury Habrard (2) , François Laviolette (3) , Emilie Morvant (2)

1 SIERRA - Statistical Machine Learning and Parsimony
2 LHC - Laboratoire Hubert Curien
3 Département d'informatique et de génie logiciel [Québec]

In this paper, 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 target distribution. On the one hand, we propose an improvement of the previous approach proposed by Germain et al. (2013), that relies on a novel distribution pseudodistance based on a disagreement averaging, allowing us to derive a new tighter PAC-Bayesian domain adaptation bound for the stochastic Gibbs classifier. 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. On the other hand, we generalize these results to multisource domain adaptation allowing us to take into account different source domains. This study opens the door to tackle domain adaptation tasks by making use of all the PAC-Bayesian tools.

Mots clés: Domain Adaptation; PAC-Bayesian Theory; Multisource

Domaines: Machine Learning [stat.ML]

Pascal Germain, Amaury Habrard, François Laviolette, Emilie Morvant. PAC-Bayesian Theorems for Domain Adaptation with Specialization to Linear Classifiers. [Research Report] Université Jean Monnet, Saint-Étienne (42); Département d'Informatique et de Génie Logiciel, Université Laval (Québec); ENS Paris; IST Austria. 2016. ⟨hal-01134246v3⟩