Using Latent Codes for Class Imbalance Problem in Unsupervised Domain Adaptation
This addresses a practical problem in real-world domain adaptation where class imbalance is common, but it appears incremental as it builds on existing adversarial frameworks.
The paper tackles severe class imbalance in unsupervised domain adaptation where source and target class spaces diverge, proposing a method using latent codes to disentangle target domain structure and identify under-represented classes, resulting in accurate target label estimation.
We address the problem of severe class imbalance in unsupervised domain adaptation, when the class spaces in source and target domains diverge considerably. Till recently, domain adaptation methods assumed the aligned class spaces, such that reducing distribution divergence makes the transfer between domains easier. Such an alignment assumption is invalidated in real world scenarios where some source classes are often under-represented or simply absent in the target domain. We revise the current approaches to class imbalance and propose a new one that uses latent codes in the adversarial domain adaptation framework. We show how the latent codes can be used to disentangle the silent structure of the target domain and to identify under-represented classes. We show how to learn the latent code reconstruction jointly with the domain invariant representation and use them to accurately estimate the target labels.