Adaptive contrastive learning with label consistency for source data free unsupervised domain adaptation
Authors:
- Xuejun Zhao,
- Rafal Stanislawski,
- Paolo Gardoni,
- Maciej Sulowicz,
- Adam Glowacz,
- Grzegorz Krolczyk,
- Zhixiong Li
Abstract
Unsupervised domain adaptation, which aims to alleviate the domain shift between source domain and target domain, has attracted extensive research interest; however, this is unlikely in practical application scenarios, which may be due to privacy issues and intellectual rights. In this paper, we discuss a more challenging and practical source-free unsupervised domain adaptation, which needs to adapt the source domain model to the target domain without the aid of source domain data. We propose label consistent contrastive learning (LCCL), an adaptive contrastive learning framework for source-free unsupervised domain adaptation, which encourages target domain samples to learn class-level discriminative features. Considering that the data in the source domain are unavailable, we introduce the memory bank to store the samples with the same pseudo label output and the samples obtained by clustering, and the trusted historical samples are involved in contrastive learning. In addition, we demonstrate that LCCL is a general framework that can be applied to unsupervised domain adaptation. Extensive experiments on digit recognition and image classification benchmark datasets demonstrate the effectiveness of the proposed method.
- Record ID
- CUT79c80af03a8f441688b8960ae259ad61
- Publication categories
- ;
- Author
- Journal series
- Sensors, ISSN , e-ISSN 1424-8220, Biweekly
- Issue year
- 2022
- Vol
- 22
- No
- 11
- Pages
- [1-13]
- Article number
- 4238
- Other elements of collation
- rys.; schem.; tab.; wykr.; Bibliografia (na s.) - 11-13; Bibliografia (liczba pozycji) - 44; Oznaczenie streszczenia - Abstr.; Numeracja w czasopiśmie - Vol. 22, Iss. 11, Spec. Iss.
- Substantive notes
- Special Issue: Artificial Intelligence for Fault Diagnostics and Prognostics
- Keywords in English
- unsupervised domain adaptation, contrastive learning, source free domain adaptation
- ASJC Classification
- ; ; ;
- DOI
- DOI:10.3390/s22114238 Opening in a new tab
- URL
- https://www.mdpi.com/1424-8220/22/11/4238 Opening in a new tab
- Related project
- Nowatorska, sterowana danymi, oparta na inteligentnym prognozowaniu platforma do złożonych systemów cyber-fizycznych w kierunku przyszłości. . Project leader at PK: , ,
- Language
- eng (en) English
- License
- Score (nominal)
- 100
- Score source
- journalList
- Score
- Publication indicators
- Citation count
- 8
- Additional fields
- Indeksowana w: Web of Science, Scopus
- Uniform Resource Identifier
- https://cris.pk.edu.pl/info/article/CUT79c80af03a8f441688b8960ae259ad61/
- URN
urn:pkr-prod:CUT79c80af03a8f441688b8960ae259ad61
* presented citation count is obtained through Internet information analysis, and it is close to the number calculated by the Publish or PerishOpening in a new tab system.