Towards Fair Cross-Domain Adaptation via Generative Learning
This work addresses fairness in cross-domain adaptation for real-world applications where data collection is difficult, though it is incremental as it builds on existing domain adaptation methods.
The paper tackles the problem of domain adaptation when source classes are imbalanced, leading to poor performance on minority categories in the target domain, and proposes a generative method that improves few-shot and overall classification accuracy on two visual datasets compared to state-of-the-art approaches.
Domain Adaptation (DA) targets at adapting a model trained over the well-labeled source domain to the unlabeled target domain lying in different distributions. Existing DA normally assumes the well-labeled source domain is class-wise balanced, which means the size per source class is relatively similar. However, in real-world applications, labeled samples for some categories in the source domain could be extremely few due to the difficulty of data collection and annotation, which leads to decreasing performance over target domain on those few-shot categories. To perform fair cross-domain adaptation and boost the performance on these minority categories, we develop a novel Generative Few-shot Cross-domain Adaptation (GFCA) algorithm for fair cross-domain classification. Specifically, generative feature augmentation is explored to synthesize effective training data for few-shot source classes, while effective cross-domain alignment aims to adapt knowledge from source to facilitate the target learning. Experimental results on two large cross-domain visual datasets demonstrate the effectiveness of our proposed method on improving both few-shot and overall classification accuracy comparing with the state-of-the-art DA approaches.