2020/06/23 by Rosanna Turrisi, Turrisi, Rosanna, Rémi Flamary +5 · 1 citation
Computer Science · Engineering · Mathematics · #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Multimodal Machine Learning Applications #Water Systems and Optimization #cs.LG #stat.ML
paper · pdf · doi:10.48550/arxiv.2006.12938
Accepted at UAI 2022
openalex publication_date 2020/06/23 · arxiv created 2022/06/02 · arxiv updated 2022/06/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
The problem of domain adaptation on an unlabeled target dataset using knowledge from multiple labelled source datasets is becoming increasingly important. A key challenge is to design an approach that overcomes the covariate and target shift both among the sources, and between the source and target domains. In this paper, we address this problem from a new perspective: instead of looking for a latent representation invariant between source and target domains, we exploit the diversity of source distributions by tuning their weights to the target task at hand. Our method, named Weighted Joint Distribution Optimal Transport (WJDOT), aims at finding simultaneously an Optimal Transport-based alignment between the source and target distributions and a re-weighting of the sources distributions. We discuss the theoretical aspects of the method and propose a conceptually simple algorithm. Numerical experiments indicate that the proposed method achieves state-of-the-art performance on simulated and real-life datasets.