Unbalanced minibatch Optimal Transport; applications to Domain Adaptation - Département de mathématiques appliquées Access content directly
Conference Papers Year : 2021

Unbalanced minibatch Optimal Transport; applications to Domain Adaptation

Abstract

Optimal transport distances have found many applications in machine learning for their capacity to compare non-parametric probability distributions. Yet their algorithmic complexity generally prevents their direct use on large scale datasets. Among the possible strategies to alleviate this issue, practitioners can rely on computing estimates of these distances over subsets of data, {\em i.e.} minibatches. While computationally appealing, we highlight in this paper some limits of this strategy, arguing it can lead to undesirable smoothing effects. As an alternative, we suggest that the same minibatch strategy coupled with unbalanced optimal transport can yield more robust behavior. We discuss the associated theoretical properties, such as unbiased estimators, existence of gradients and concentration bounds. Our experimental study shows that in challenging problems associated to domain adaptation, the use of unbalanced optimal transport leads to significantly better results, competing with or surpassing recent baselines.
Fichier principal
Vignette du fichier
2103.03606v1.pdf (2.56 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03264020 , version 1 (12-07-2024)

Identifiers

Cite

Kilian Fatras, Thibault Séjourné, Nicolas Courty, Rémi Flamary. Unbalanced minibatch Optimal Transport; applications to Domain Adaptation. International Conference in machine Learning, Jul 2021, online, France. ⟨10.48550/arXiv.2103.03606⟩. ⟨hal-03264020⟩
73 View
0 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More