التفاصيل البيبلوغرافية
| العنوان: |
Uncertainty-guided Source-free Domain Adaptation |
| المؤلفون: |
Subhankar Roy, Martin Trapp, Andrea Pilzer, Juho Kannala, Nicu Sebe, Elisa Ricci, Arno Solin |
| المصدر: |
European Conference on Computer Vision 2022 |
| بيانات النشر: |
Zenodo |
| سنة النشر: |
2023 |
| المجموعة: |
Zenodo |
| الوصف: |
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. |
| نوع الوثيقة: |
conference object |
| اللغة: |
English |
| Relation: |
https://zenodo.org/communities/ai4media; https://doi.org/10.5281/zenodo.7566108; https://doi.org/10.5281/zenodo.7566109; oai:zenodo.org:7566109 |
| DOI: |
10.5281/zenodo.7566109 |
| الاتاحة: |
https://doi.org/10.5281/zenodo.7566109 |
| Rights: |
info:eu-repo/semantics/openAccess ; Creative Commons Attribution 4.0 International ; https://creativecommons.org/licenses/by/4.0/legalcode |
| رقم الانضمام: |
edsbas.37EBD7C7 |
| قاعدة البيانات: |
BASE |