Uncertainty-guided source-free domain adaptation

Subhankar Roy*, Martin Trapp, Andrea Pilzer, Juho Kannala, Nicu Sebe, Elisa Ricci, Arno Solin

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

18 Citations (Scopus)
92 Downloads (Pure)


Source-free domain adaptation (SFDA) aims to adapt a classifier to an unlabelled target data set by only using a pre-trained source model. However, the absence of the source data and the domain shift makes the predictions on the target data unreliable. We propose quantifying the uncertainty in the source model predictions and utilizing it to guide the target adaptation. For this, we construct a probabilistic source model by incorporating priors on the network parameters inducing a distribution over the model predictions. Uncertainties are estimated by employing a Laplace approximation and incorporated to identify target data points that do not lie in the source manifold and to down-weight them when maximizing the mutual information on the target data. Unlike recent works, our probabilistic treatment is computationally lightweight, decouples source training and target adaptation, and requires no specialized source training or changes of the model architecture. We show the advantages of uncertainty-guided SFDA over traditional SFDA in the closed-set and open-set settings and provide empirical evidence that our approach is more robust to strong domain shifts even without tuning.
Original languageEnglish
Title of host publicationComputer Vision – ECCV 2022
Subtitle of host publication17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part XXV
EditorsShai Avidan, Gabriel Brostow, Moustapha Cissé, Giovanni Maria Farinella, Tal Hassner
ISBN (Electronic)978-3-031-19806-9
ISBN (Print)978-3-031-19805-2
Publication statusPublished - 2022
MoE publication typeA4 Conference publication
EventEuropean Conference on Computer Vision - Tel Aviv, Israel
Duration: 23 Oct 202227 Oct 2022
Conference number: 17

Publication series

NameLecture Notes in Computer Science
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


ConferenceEuropean Conference on Computer Vision
Abbreviated titleECCV
CityTel Aviv
Internet address


Dive into the research topics of 'Uncertainty-guided source-free domain adaptation'. Together they form a unique fingerprint.

Cite this