Unpaired, unsupervised domain adaptation assumes your domains are already similar

Gijs van Tulder*, Marleen de Bruijne

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

1 Citation (Scopus)
27 Downloads (Pure)

Abstract

Unsupervised domain adaptation is a popular method in medical image analysis, but it can be tricky to make it work: without labels to link the domains, domains must be matched using feature distributions. If there is no additional information, this often leaves a choice between multiple possibilities to map the data that may be equally likely but not equally correct. In this paper we explore the fundamental problems that may arise in unsupervised domain adaptation, and discuss conditions that might still make it work. Focusing on medical image analysis, we argue that images from different domains may have similar class balance, similar intensities, similar spatial structure, or similar textures. We demonstrate how these implicit conditions can affect domain adaptation performance in experiments with synthetic data, MNIST digits, and medical images. We observe that practical success of unsupervised domain adaptation relies on existing similarities in the data, and is anything but guaranteed in the general case. Understanding these implicit assumptions is a key step in identifying potential problems in domain adaptation and improving the reliability of the results.

Original languageEnglish
Article number102825
JournalMedical Image Analysis
Volume87
DOIs
Publication statusPublished - Jul 2023

Bibliographical note

Funding Information:
The authors received funding from the Dutch Research Council (NWO) with grant numbers 639.022.010 and VI.C.182.042 .

Publisher Copyright: © 2023 The Author(s)

Fingerprint

Dive into the research topics of 'Unpaired, unsupervised domain adaptation assumes your domains are already similar'. Together they form a unique fingerprint.

Cite this