App. Res. of Comp. Accepted Paper
Vol. 44, 2027 No. 2

Dual-adaptive differential privacy federated learning method for client heterogeneity

Liu Yi1
Li Xinyi1
Chen Xinzhuang1
Yan Duli2
Wang Wenzhuo1
1. College of Mathematics and Computer Science, Yan'an University, Yan'an 716000, China
2. School of Artificial Intelligence and Computer Science, Shaanxi Normal University, Xi'an 710062, China
DOI: 10.19734/j.issn.1001-3695.2026.06.0178

Abstract

To address client heterogeneity in federated learning, this paper proposes a client-aware dual adaptive differential privacy framework (CADA-DP) . In federated learning, clients often differ in data size, data distribution, and gradient characteristics. Traditional differential privacy methods usually rely on fixed noise injection and static privacy budget allocation. As a result, they often fail to achieve a good balance between model utility and privacy protection. CADA-DP combines a gradient-aware adaptive Gaussian noise mechanism with a client-aware privacy budget allocation strategy. It dynamically adjusts the noise scale and privacy budget according to the historical training states of the clients, which helps reduce the training instability caused by heterogeneous data distributions. In each training round, the system uses the clipped and Gaussian-perturbed updates uploaded in the previous round to estimate gradient changes through an exponential moving average, and accordingly adjusts the noise scale and privacy budget for the current round. The clients then clip and perturb their local updates before uploading them to the server for aggregation. Experimental results on MNIST, FMNIST, and CIFAR-10 show that, under unified training settings and RDP privacy accounting, CADA-DP achieves a lower cumulative privacy loss and higher classification accuracy. Under entropy-based membership inference attacks, the attack metrics are close to random guessing. The framework also shows good stability and robustness under different hyperparameter settings and experimental configurations.

Foundation Support

国家自然科学基金资助项目(12561065)
陕西省教育厅一般专项(24JK0723)
延安市青年人才托举(YAU202506703)

Publish Information

DOI: 10.19734/j.issn.1001-3695.2026.06.0178
Publish at: Application Research of Computers Accepted Paper, Vol. 44, 2027 No. 2

Publish History

[2026-10-23] Accepted Paper

Cite This Article

刘翼, 李欣怡, 陈新庄, 等. 面向客户端异构性的双重自适应差分隐私联邦学习方法 [J]. 计算机应用研究, 2027, 44 (2). (2026-10-24). https://doi.org/10.19734/j.issn.1001-3695.2026.06.0178. (Liu Yi, Li Xinyi, Chen Xinzhuang, et al. Dual-adaptive differential privacy federated learning method for client heterogeneity [J]. Application Research of Computers, 2027, 44 (2). (2026-10-24). https://doi.org/10.19734/j.issn.1001-3695.2026.06.0178. )

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
  • 2020-2025 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)