App. Res. of Comp. Accepted Paper
Vol. 44, 2027 No. 1

Personalized federated learning with global prototype relation awareness and selective sharpness optimization

Tan Yiting1
Li Xuanchi1
Wen Jing2
Liang Shaoling2
Huang Baohua1
1. School of Computer and Electronic Information, Guangxi University, Nanning Guangxi 530004, China
2. Guangxi Information Center, Nanning 530000, China
DOI: 10.19734/j.issn.1001-3695.2026.04.0138

Abstract

To address prototype bias under heterogeneous data scenarios and local model overfitting in personalized federated learning, this paper proposes FedGSO (Personalized Federated Learning with Global Prototype Relation Awareness and Selective Sharpness Optimization) . The framework parameterizes local class prototypes and utilizes the similarity relationships among global prototypes to dynamically adjust contrastive learning constraints. This operation separates easily confused classes in the feature space. Furthermore, the proposed method introduces a selective layer-wise sharpness-aware minimization strategy. This strategy applies perturbations only to key shared representation layers to guide the model towards flat minima. Experiments on three public datasets demonstrate the superiority of FedGSO over recent personalized federated learning methods. Specifically, under two typical heterogeneous settings on CIFAR-100, FedGSO achieved test accuracies of 72.86% and 62.06%, improving upon the best baseline by 2.19% and 2.70%, respectively. These results verify the effectiveness of this method for separating easily confused classes and enhancing model generalization under heterogeneous data.

Foundation Support

广西数字基础设施重点实验室开放基金(GXDINBC202406)
国家自然科学基金(61962005)

Publish Information

DOI: 10.19734/j.issn.1001-3695.2026.04.0138
Publish at: Application Research of Computers Accepted Paper, Vol. 44, 2027 No. 1

Publish History

[2026-08-28] Accepted Paper

Cite This Article

谭以婷, 李炫池, 文静, 等. 全局原型关系感知与选择性锐度优化的个性化联邦学习 [J]. 计算机应用研究, 2027, 44 (1). (2026-09-14). https://doi.org/10.19734/j.issn.1001-3695.2026.04.0138. (Tan Yiting, Li Xuanchi, Wen Jing, et al. Personalized federated learning with global prototype relation awareness and selective sharpness optimization [J]. Application Research of Computers, 2027, 44 (1). (2026-09-14). https://doi.org/10.19734/j.issn.1001-3695.2026.04.0138. )

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)