Informative Feature Disentanglement for Unsupervised Domain Adaptation

Unsupervised Domain Adaptation (UDA) aims at learning a classifier for an unlabeled target domain by transferring knowledge from a labeled source domain with a related but different distribution. The strategy of aligning the two domains in latent feature space via metric discrepancy or adversarial learning has achieved much progress. However, these existing approaches mainly focus on the adaptation of the entire image, ignoring the bottleneck that forcing adaptation of uninformative domain-specific variations may undermine the effectiveness of learned features. To address this problem, we propose a novel component called Informative Feature Disentanglement (IFD) which is equipped with the adversarial network or the metric discrepancy model, respectively. Accordingly, the new network architectures, named IFDAN and IFDMN, enable informative features refinement before the adaptation. The proposed IFD is designed to disentangle informative features from the uninformative domain-specific variations, which is a Variational Autoencoder (VAE) with lateral connections from the encoder to the decoder. We apply the IFD to conduct supervised disentanglement for the source domain and unsupervised disentanglement for the target domain cooperatively. In this way, informative features are disentangled from the domain-specific details before the adaptation. Extensive experimental results on three gold-standard domain adaptation datasets, e.g., Office31, Office-Home and VisDA-C, demonstrate the effectiveness of the proposed model IFDAN and IFDMN for UDA.