University of Oulu

Deng, W., Su, Z., Qiu, Q., Zhao, L., Kuang, G., Pietikäinen, M., Xiao, H., & Liu, L. (2021). Deep ladder reconstruction-classification network for unsupervised domain adaptation. Pattern Recognition Letters, 152, 398–405. https://doi.org/10.1016/j.patrec.2021.10.009

Deep ladder reconstruction-classification network for unsupervised domain adaptation

Saved in:
Author: Deng, Wanxia1; Su, Zhuo2; Qiu, Qiang3;
Organizations: 1State Key Laboratory of Complex Electromagnetic Environment Effects on Electronics and Information System, College of Electronic Science, National University of Defense technology, Changsha, Hunan 410003, China
2University of Oulu, Oulu 90014, Finland
3Duke University, Durham 27708, USA
4College of Systems Engineering, National University of Defense technology, Changsha, Hunan 410003, China
Format: article
Version: accepted version
Access: embargoed
Persistent link: http://urn.fi/urn:nbn:fi-fe2022022520822
Language: English
Published: Elsevier, 2021
Publish Date: 2023-10-12
Description:

Abstract

Unsupervised Domain Adaptation aims to learn a classifier for an unlabeled target domain by transferring knowledge from a labeled source domain. Most existing approaches learn domain-invariant features by adapting the entire information of each image. However, forcing adaptation of domain-specific components can undermine the effectiveness of learned features. We propose a novel architecture called Deep Ladder Reconstruction-Classification Network (DLaReC) which is designed to learn cross-domain shared contents by suppressing domain-specific variations. The DLaReC adopts an encoder with cross-domain sharing and a target-domain reconstruction decoder. The encoder and decoder are connected with residual shortcuts at each intermediate layer. By this means, the domain-specific components are directly fed to the decoder for reconstruction, relieving the pressure to learn domain-specific variations at later layers of the shared encoder. Therefore, DLaReC allows the encoder to focus on learning cross-domain shared representations and ignore domain-specific variations. DLaReC is implemented by jointly learning three tasks: supervised classification of the source domain, unsupervised reconstruction of the target domain and cross-domain shared representation adaptation. Extensive experiments on Digit, Office31, ImageCLEF-DA and Office-Home datasets demonstrate the DLaReC outperforms state-of-the-art methods on the whole. The average accuracy on the Digit datasets, for instance, is improved from 95.6% to 96.9%. In addition, the result on Amazon → Webcam obtains significant improvement, i.e., from 91.1% to 94.7%.

see all

Series: Pattern recognition letters
ISSN: 0167-8655
ISSN-E: 1872-7344
ISSN-L: 0167-8655
Volume: 152
Pages: 398 - 405
DOI: 10.1016/j.patrec.2021.10.009
OADOI: https://oadoi.org/10.1016/j.patrec.2021.10.009
Type of Publication: A1 Journal article – refereed
Field of Science: 113 Computer and information sciences
Subjects:
Funding: This work was supported by the National Natural Science Foundation of China under Grant 61701508, and Hunan Provincial Natural Science Foundation of China under Grant 2018JJ3613, and Science and Technology on Electro-optic Control Laboratory under Grant 20165188004 and China Scholarship Council.
Copyright information: © 2021 Published by Elsevier B.V. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/.
  https://creativecommons.org/licenses/by-nc-nd/4.0/