Generalized Domain Adaptation
This work addresses the practical limitation of domain adaptation methods being ineffective across different variants, which hinders real-world applications, by providing a generalized framework and solution.
The paper tackles the lack of a unified framework for unsupervised domain adaptation (UDA) variants by introducing Generalized Domain Adaptation (GDA), which organizes major variants and addresses a new challenging setting where domain labels are unknown and class labels are partially given. The proposed method, using self-supervised class-destructive learning, outperforms state-of-the-art UDA methods in this new setting and is competitive in existing variations, as demonstrated by extensive experiments on three benchmark datasets.
Many variants of unsupervised domain adaptation (UDA) problems have been proposed and solved individually. Its side effect is that a method that works for one variant is often ineffective for or not even applicable to another, which has prevented practical applications. In this paper, we give a general representation of UDA problems, named Generalized Domain Adaptation (GDA). GDA covers the major variants as special cases, which allows us to organize them in a comprehensive framework. Moreover, this generalization leads to a new challenging setting where existing methods fail, such as when domain labels are unknown, and class labels are only partially given to each domain. We propose a novel approach to the new setting. The key to our approach is self-supervised class-destructive learning, which enables the learning of class-invariant representations and domain-adversarial classifiers without using any domain labels. Extensive experiments using three benchmark datasets demonstrate that our method outperforms the state-of-the-art UDA methods in the new setting and that it is competitive in existing UDA variations as well.