Pré-Publication, Document De Travail Année : 2025

Bridging Contrastive Learning and Domain Adaptation: Theoretical Perspective and Practical Application

Résumé

This work studies the relationship between Contrastive Learning and Domain Adaptation from a theoretical perspective. The two standard contrastive losses, NT-Xent loss (Self-supervised) and Supervised Contrastive loss, are related to the Class-wise Mean Maximum Discrepancy (CMMD), a dissimilarity measure widely used for Domain Adaptation. Our work shows that minimizing the contrastive losses decreases the CMMD and simultaneously improves classseparability, laying the theoretical groundwork for the use of Contrastive Learning in the context of Domain Adaptation. Due to the relevance of Domain Adaptation in medical imaging, we focused the experiments on mammography images. Extensive experiments on three mammography datasets -synthetic patches, clinical (real) patches, and clinical (real) images -show improved Domain Adaptation, class-separability, and classification performance, when minimizing the Supervised Contrastive loss.
Fichier principal
Vignette du fichier
icml2025_preprint (1).pdf (7) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
licence

Dates et versions

hal-04922401 , version 1 (30-01-2025)

Licence

Identifiants

  • HAL Id : hal-04922401 , version 1

Citer

Gonzalo Iñaki Quintana, Laurence Vancamberg, Vincent Jugnon, Agnès Desolneux, Mathilde Mougeot. Bridging Contrastive Learning and Domain Adaptation: Theoretical Perspective and Practical Application. 2025. ⟨hal-04922401⟩
0 Consultations
0 Téléchargements

Partager

More