Analyse théorique de l’apprentissage avec des fonctions de similarités pour l’adaptation de domaine

Sofien Dhouib (1) , Ievgen Redko (1)

1 Images et Modèles

Similarity learning is an active research area in machine learning that tackles the problem of finding a similarity function tailored to an observable data sample in order to achieve efficient classification. This learning scenario has been generally formalized by the means of a (, γ, τ)−good similarity learning framework in the context of supervised classification and has been shown to have important theoretical guarantees. In this paper , we propose to extend the theoretical analysis of similarity learning to the domain adaptation setting, a particular situation occurring when the similarity is learned and then deployed on samples following different probability distributions. We give a new definition of an (, γ)−good similarity for domain adaptation and prove several results quantifying the performance of a similarity function on a target domain after it has been trained on a source domain. We particularly show that if the source domain support contains that of the target then a notable improvement of the adaptation is achievable.

Mots clés: domain adaptation; similarity learning; metric learning

Domaines: Probabilités [math.PR]; Intelligence artificielle [cs.AI]

Sofien Dhouib, Ievgen Redko. Analyse théorique de l’apprentissage avec des fonctions de similarités pour l’adaptation de domaine. Conférence sur l'Apprentissage Automatique 2018, Jun 2018, Rouen, France. ⟨hal-02063285⟩