Unsupervised Domain Adaptation via Regularized Conditional Alignment

Safa Cicek, Stefano Soatto; The IEEE International Conference on Computer Vision (ICCV), 2019, pp. 1416-1425

Abstract


We propose a method for unsupervised domain adaptation that trains a shared embedding to align the joint distributions of inputs (domain) and outputs (classes), making any classifier agnostic to the domain. Joint alignment ensures that not only the marginal distributions of the domains are aligned, but the labels as well. We propose a novel objective function that encourages the class-conditional distributions to have disjoint support in feature space. We further exploit adversarial regularization to improve the performance of the classifier on the domain for which no annotated data is available.

Related Material


[pdf] [supp]
[bibtex]
@InProceedings{Cicek_2019_ICCV,
author = {Cicek, Safa and Soatto, Stefano},
title = {Unsupervised Domain Adaptation via Regularized Conditional Alignment},
booktitle = {The IEEE International Conference on Computer Vision (ICCV)},
month = {October},
year = {2019}
}