Using Latent Codes for Class Imbalance Problem in Unsupervised Domain Adaptation
arXiv:1909.08962
Abstract
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.
References in corpus (9)
- Conditional Generative Adversarial Nets
- How transferable are features in deep neural networks?
- Deep Visual Domain Adaptation: A Survey
- Conditional Adversarial Domain Adaptation
- InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
- Unsupervised Domain Adaptation with Residual Transfer Networks
- Learning to Reweight Examples for Robust Deep Learning
- VisDA: The Visual Domain Adaptation Challenge
- Learning to Transfer Examples for Partial Domain Adaptation