Class Subset Selection for Partial Domain Adaptation

Fariba Zohrizadeh, Mohsen Kheirandishfard, Farhad Kamangar; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2019, pp. 66-72

Domain adaptation is the task of transferring knowledge from a labeled source dataset to an unlabeled target dataset. Partial domain adaptation (PDA) investigates the scenarios in which the target label space is a subset of the source label space. The main purpose of the PDA is to identify the shared classes between the domains and promote learning transferable knowledge from these classes. Inspired by the idea of subset selection, we propose an adversarial PDA approach which aims to not only automatically select the most relevant subset of source domain classes but also ignore the samples that are less transferable across the domains. In the absence of target labels, the proposed approach is able to effectively learn domain-invariant feature representations, which in turn can facilitate and enhance the classification performance in the target domain. Empirical results on Office-31 and Office-Home datasets demonstrate the high potential of the proposed approach in addressing different partial domain adaptation tasks.