Optimal Transport for Conditional Domain Matching and Label Shift - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE Accéder directement au contenu
Article Dans Une Revue Machine Learning Année : 2022

Optimal Transport for Conditional Domain Matching and Label Shift

Résumé

We address the problem of unsupervised domain adaptation under the setting of generalized target shift (both class-conditional and label shifts occur). We show that in that setting, for good generalization, it is necessary to learn with similar source and target label distributions and to match the class-conditional probabilities. For this purpose, we propose an estimation of target label proportion by blending mixture estimation and optimal transport. This estimation comes with theoretical guarantees of correctness. Based on the estimation, we learn a model by minimizing a importance weighted loss and a Wasserstein distance between weighted marginals. We prove that this minimization allows to match class-conditionals given mild assumptions on their geometry. Our experimental results show that our method performs better on average than competitors accross a range domain adaptation problems including digits,VisDA and Office. Code for this paper is available at \url{https://github.com/arakotom/mars_domain_adaptation}.
Fichier principal
Vignette du fichier
dann_targetshift_arxiv.pdf (1.78 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-02866979 , version 1 (12-06-2020)
hal-02866979 , version 2 (15-10-2020)

Identifiants

Citer

Alain Rakotomamonjy, Rémi Flamary, Gilles Gasso, M. El Alaya, Maxime Berar, et al.. Optimal Transport for Conditional Domain Matching and Label Shift. Machine Learning, 2022, 111 (5), pp.1651-1670. ⟨10.1007/s10994-021-06088-2⟩. ⟨hal-02866979v2⟩
342 Consultations
168 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More