Conditional Adversarial Domain Adaptation

Authors:
Mingsheng Long Tsinghua University
Zhangjie CAO Stanford University
Jianmin Wang Tsinghua University
Michael I. Jordan UC Berkeley

Introduction:

Adversarial learning has been embedded into deep networks to learn disentangled and transferable representations for domain adaptation.

Abstract:

Adversarial learning has been embedded into deep networks to learn disentangled and transferable representations for domain adaptation. Existing adversarial domain adaptation methods may struggle to align different domains of multimodal distributions that are native in classification problems. In this paper, we present conditional adversarial domain adaptation, a principled framework that conditions the adversarial adaptation models on discriminative information conveyed in the classifier predictions. Conditional domain adversarial networks (CDANs) are designed with two novel conditioning strategies: multilinear conditioning that captures the cross-covariance between feature representations and classifier predictions to improve the discriminability, and entropy conditioning that controls the uncertainty of classifier predictions to guarantee the transferability. Experiments testify that the proposed approach exceeds the state-of-the-art results on five benchmark datasets.

You may want to know: