Technology of Information Security
|
App. Res. of Comp. Printed Article
Vol. 43, 2026 No. 1
256-262

Backdoor defense via self-supervised learning and dataset splitting

He Zisheng
Ling Jie
School of Computer Science and Technology, Guangdong University of Technology, Guangzhou 510006, China
DOI: 10.19734/j.issn.1001-3695.2025.05.0190

Abstract

To address the vulnerability of deep neural networks(DNNs) to backdoor attacks in image classification and the challenge of balancing model accuracy and robustness in existing defenses, this paper proposed a semi-supervised backdoor defense method named SAS, which was based on self-supervised pre-training and dynamic dataset splitting. The method firstly employed a self-supervised training phase using a contrastive learning framework with consistency regularization to decouple image features from backdoor patterns. Subsequently, the fine-tuning stage utilized a strategy combining dynamic data selection and semi-supervised learning. This strategy identified and leveraged high-confidence and low-confidence data separately during training to suppress backdoor implantation. Experiments on the CIFAR-10 and GTSRB datasets against four attacks(BadNets, Blend, WaNet, and Refool) demonstrated that the proposed method, compared to the ASD method, improved the classification accuracy on clean data by an average of 1.65 and 0.65 percent points, respectively. Furthermore, it consistently reduced the backdoor attack success rate on poisoned data to below 1.4%. The results confirm that the synergy between feature decoupling and dynamic dataset splitting enables this method to effectively enhance the model's backdoor robustness while maintaining high performance on clean data, providing an effective pathway for building secure and reliable deep learning models.

Foundation Support

广州市重点领域研发计划项目(202007010004)
广州开发区国际科技合作项目(2023GH05)

Publish Information

DOI: 10.19734/j.issn.1001-3695.2025.05.0190
Publish at: Application Research of Computers Printed Article, Vol. 43, 2026 No. 1
Section: Technology of Information Security
Pages: 256-262
Serial Number: 1001-3695(2026)01-031-0256-07

Publish History

[2025-09-15] Accepted Paper
[2026-01-05] Printed Article

Cite This Article

何子晟, 凌捷. 基于自监督学习与数据集分割的后门防御方法 [J]. 计算机应用研究, 2026, 43 (1): 256-262. (He Zisheng, Ling Jie. Backdoor defense via self-supervised learning and dataset splitting [J]. Application Research of Computers, 2026, 43 (1): 256-262. )

About the Journal

  • Application Research of Computers Monthly Journal
  • Journal ID ISSN 1001-3695
    CN  51-1196/TP

Application Research of Computers, founded in 1984, is an academic journal of computing technology sponsored by Sichuan Institute of Computer Sciences under the Science and Technology Department of Sichuan Province.

Aiming at the urgently needed cutting-edge technology in this discipline, Application Research of Computers reflects the mainstream technology, hot technology and the latest development trend of computer application research at home and abroad in a timely manner. The main contents of the journal include high-level academic papers in this discipline, the latest scientific research results and major application results. The contents of the columns involve new theories of computer discipline, basic computer theory, algorithm theory research, algorithm design and analysis, blockchain technology, system software and software engineering technology, pattern recognition and artificial intelligence, architecture, advanced computing, parallel processing, database technology, computer network and communication technology, information security technology, computer image graphics and its latest hot application technology.

Application Research of Computers has many high-level readers and authors, and its readers are mainly senior and middle-level researchers and engineers engaged in the field of computer science, as well as teachers and students majoring in computer science and related majors in colleges and universities. Over the years, the total citation frequency and Web download rate of Application Research of Computers have been ranked among the top of similar academic journals in this discipline, and the academic papers published are highly popular among the readers for their novelty, academics, foresight, orientation and practicality.

Indexed & Evaluation

  • The Second National Periodical Award 100 Key Journals
  • Double Effect Journal of China Journal Formation
  • the Core Journal of China (Peking University 2023 Edition)
  • the Core Journal for Science
  • Chinese Science Citation Database (CSCD) Source Journals
  • RCCSE Chinese Core Academic Journals
  • Journal of China Computer Federation
  • 2020-2022 The World Journal Clout Index (WJCI) Report of Scientific and Technological Periodicals
  • Full-text Source Journal of China Science and Technology Periodicals Database
  • Source Journal of China Academic Journals Comprehensive Evaluation Database
  • Source Journals of China Academic Journals (CD-ROM Version), China Journal Network
  • 2017-2019 China Outstanding Academic Journals with International Influence (Natural Science and Engineering Technology)
  • Source Journal of Top Academic Papers (F5000) Program of China's Excellent Science and Technology Journals
  • Source Journal of China Engineering Technology Electronic Information Network and Electronic Technology Literature Database
  • Source Journal of British Science Digest (INSPEC)
  • Japan Science and Technology Agency (JST) Source Journal
  • Russian Journal of Abstracts (AJ, VINITI) Source Journals
  • Full-text Journal of EBSCO, USA
  • Cambridge Scientific Abstracts (Natural Sciences) (CSA(NS)) core journals
  • Poland Copernicus Index (IC)
  • Ulrichsweb (USA)