Visual domain adaptation based on modified A−distance and sparse filtering. (August 2020)

• Record Type: Journal Article
• Authors: Han, Chao; Lei, Yu; Xie, Yu; Zhou, Deyun; Gong, Maoguo
• Is Part Of: Pattern recognition. Volume 104(2020:Aug.)
• Publication Date: 2020-08
• Subjects: Domain adaptation -- A−distance -- Sparse filtering; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2020.107254
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: A novel model is proposed for domain adaptation, which does not require any labels from source or target domain during training. We modify A-distance by solving the linear classification problem with pseudo inverse. We use the learning paradigm of sparse filtering to preserve the structure of data, and point out that within-domain normalization is more suitable for domain adaptation problems. Abstract: Domain adaptation studies how to build a robust model to solve pattern recognition problems when training in a source domain while testing in a related but different target domain. The existing methods focus on how to shorten the distance between the two domains, however, they have limited considerations on the preservation of data structures. In this paper, we propose a novel model for unsupervised domain adaptation. For the reduction of domain discrepancy, we propose modified A − distance, which is computationally fast and can be optimized using gradient information. Moreover, in order to capture the internal structures of target samples, within-domain normalization based sparse filtering is raised, which proved to be more powerful for domain adaptation. Extensive experiments compared to both shallow and deep methods demonstrate the effectiveness of our approach.