Class Discriminative Adversarial Learning for Unsupervised Domain Adaptation

Lihua Zhou, Mao Ye, Xiatian Zhu, Shuaifeng Li, Yiguang Liu

Abstract:

As a state-of-the-art family of Unsupervised Domain Adaptation (UDA), bi-classifier adversarial learning methods are formulated in
an adversarial (minimax) learning framework with a single feature extractor and two classifiers. Model training alternates between
two steps: (I) constraining the learning of the two classifiers to maximize the prediction discrepancy of unlabeled target domain
data, and (II) constraining the learning of the feature extractor to minimize this discrepancy. Despite being an elegant formulation,
this approach has a fundamental limitation: Maximizing and minimizing the classifier discrepancy is not class discriminative for the
target domain, finally leading to a suboptimal adapted model. To solve this problem, we propose a novel Class Discriminative Adversarial Learning (CDAL) method characterized by discovering class discrimination knowledge and leveraging this knowledge to discriminatively regulate the classifier discrepancy constraints onthe-fly. This is realized by introducing an evaluation criterion for judging each classifier’s capability and each target domain sample’s feature reorientation via objective loss reformulation. Extensive
experiments on three standard benchmarks show that our CDAL method yields new state-of-the-art performance. Our code is made
available at https://github.com/buerzlh/CDAL.