EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Federated semi-supervised learning based on feature alignment and knowledge distillation

Zhe Ding, Hao Yi, Wenrui Xie, Yuxuan Xiao, Qixu Wang, Qing Chen, Zhiguang Qin, Dajiang Chen

Frontiers in Physics · 14.01.2026 · 13

DOI: 10.3389/fphy.2025.1724537 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Physics, online erschienen am 14.01.2026. Verfasst von Zhe Ding und 7 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Recently, federated learning has been successfully applied in fields related to cyber-physical-social systems (CPSSs), owing to its ability to harness decentralized clients for training a global model while ensuring data privacy. The existing methods encounter two main obstacles, namely, the statistical distribution heterogeneity [non-independent and identically distributed (non-IID)] among clients and the scarcity of labeled data.

Methods

In this article, we propose a federated semi-supervised learning (FSSL) model under the label-at-server scenario, denoted as FedAlign, which is tailored for distributed cyber-physical-social systems. FedAlign adopts a dual knowledge distillation framework to train the global model. On the client side, FedAlign integrates contrastive learning, knowledge distillation, and pseudo-labeling technology to train local models. The goal is to ensure that global knowledge is not overlooked while enabling clients to learn local knowledge. Meanwhile, on the server side, FedAlign utilizes maximum mean discrepancy to generate a global feature space. Based on the generated feature space, FedAlign employs a knowledge distillation mechanism and supervised learning to aggregate local knowledge and update the global model.

Results

Two classic datasets, CIFAR-10 and Fashion-MNIST, are used to evaluate the performance of FedAlign. The experimental results demonstrate that FedAlign outperforms traditional federated semi-supervised learning models.

Discussion

The integration of feature alignment and knowledge enables balancing local knowledge learning and aggregation of global model. As a consequence, FedAlign enhances the adaptability and generalization ability of the global model in CPSSs.

Zitieren

APA

Ding, Z., Yi, H., Xie, W., Xiao, Y., Wang, Q., Chen, Q., Qin, Z., & Chen, D. (2026). Federated semi-supervised learning based on feature alignment and knowledge distillation. Frontiers in Physics, 13. https://doi.org/10.3389/fphy.2025.1724537

Vancouver

Ding Z, Yi H, Xie W, Xiao Y, Wang Q, Chen Q, et al. Federated semi-supervised learning based on feature alignment and knowledge distillation. Frontiers in Physics. 2026;13. doi:10.3389/fphy.2025.1724537

BibTeX
@article{Ding2026,
  author = {Ding, Zhe and Yi, Hao and Xie, Wenrui and Xiao, Yuxuan and Wang, Qixu and Chen, Qing and Qin, Zhiguang and Chen, Dajiang},
  title = {Federated semi-supervised learning based on feature alignment and knowledge distillation},
  journal = {Frontiers in Physics},
  year = {2026},
  volume = {13},
  doi = {10.3389/fphy.2025.1724537}
}
Kostenlos starten →