Towards Fair Knowledge Transfer for Imbalanced Domain Adaptation

TL;DR

This paper introduces a novel framework for fair knowledge transfer in imbalanced domain adaptation, using data augmentation and dual classifiers to improve fairness and robustness across domains.

Contribution

The paper proposes a unified framework with cross-domain mixup, dual classifiers, and prototype alignment to address fairness and domain shift in imbalanced domain adaptation.

Findings

Significantly improves accuracy over state-of-the-art models.

Effectively handles class imbalance in domain adaptation.

Achieves over 20% improvement on benchmark datasets.

Abstract

Domain adaptation (DA) becomes an up-and-coming technique to address the insufficient or no annotation issue by exploiting external source knowledge. Existing DA algorithms mainly focus on practical knowledge transfer through domain alignment. Unfortunately, they ignore the fairness issue when the auxiliary source is extremely imbalanced across different categories, which results in severe under-presented knowledge adaptation of minority source set. To this end, we propose a Towards Fair Knowledge Transfer (TFKT) framework to handle the fairness challenge in imbalanced cross-domain learning. Specifically, a novel cross-domain mixup generation is exploited to augment the minority source set with target information to enhance fairness. Moreover, dual distinct classifiers and cross-domain prototype alignment are developed to seek a more robust classifier boundary and mitigate the domain shift. Such three strategies are formulated into a unified framework to address the fairness issue and domain shift challenge. Extensive experiments over two popular benchmarks have verified the effectiveness of our proposed model by comparing to existing state-of-the-art DA models, and especially our model significantly improves over 20% on two benchmarks in terms of the overall accuracy.