EUVIMEDEUVIMED entdecken
Crossref

KAFL-HD: Knowledge Alignment in Asynchronous Federated Learning With Heterogeneous Data

Mingjian Zhi, Yuanguo Bi, Lin Cai, Tianao Xiang

IEEE Transactions on Mobile Computing · 2026-06 · 25 (6) · 8252-8267

DOI: 10.1109/tmc.2025.3647655 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Mobile Computing, 2026-06. Verfasst von Mingjian Zhi und 3 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 45 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Shenyang Science and Technology Plan Fund Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Asynchronous Federated Learning (AFL) can mitigate the straggler problem due to unbalanced training time of clients in Synchronous Federated Learning (SFL), thereby reducing the aggregation time and improving the training efficiency. However, AFL introduces training bias since different updating frequencies of heterogeneous clients can cause unequal knowledge contributions to the global model. Meanwhile, if the client data are heterogeneous, the local optimum may be drifted from the global one, which exacerbates the training bias problem. In order to solve the above issues, we propose a Knowledge Alignment framework for AFL with Heterogeneous Data, termed as KAFL-HD. Firstly, considering data heterogeneity, a data quality-aware aggregation method is proposed to estimate client contributions precisely, where both model staleness and data quality are utilized in aggregation weights. Secondly, a knowledge distillation method with staleness is designed to supplement more knowledge from slow clients to the global model. Thirdly, an adaptive learning rate adjustment method is proposed to customize the local learning rate based on the aggregation frequency and weight, which aligns the knowledge contributions of clients in the local training process. Furthermore, we provide theoretical analysis under a non-convex setting to show the convergence speed of KAFL-HD. Finally, comprehensive experiments are conducted, and the results show that KAFL-HD achieves the highest accuracy and fairness performance compared to the state-of-the-art baselines.

Förderung

  • National Natural Science Foundation of China
  • Shenyang Science and Technology Plan Fund Project

Zitieren

APA

Zhi, M., Bi, Y., Cai, L., & Xiang, T. (2026). KAFL-HD: Knowledge Alignment in Asynchronous Federated Learning With Heterogeneous Data. IEEE Transactions on Mobile Computing, 25(6), 8252-8267. https://doi.org/10.1109/tmc.2025.3647655

Vancouver

Zhi M, Bi Y, Cai L, Xiang T. KAFL-HD: Knowledge Alignment in Asynchronous Federated Learning With Heterogeneous Data. IEEE Transactions on Mobile Computing. 2026;25(6):8252-8267. doi:10.1109/tmc.2025.3647655

BibTeX
@article{Zhi2026,
  author = {Zhi, Mingjian and Bi, Yuanguo and Cai, Lin and Xiang, Tianao},
  title = {KAFL-HD: Knowledge Alignment in Asynchronous Federated Learning With Heterogeneous Data},
  journal = {IEEE Transactions on Mobile Computing},
  year = {2026},
  volume = {25},
  number = {6},
  pages = {8252--8267},
  doi = {10.1109/tmc.2025.3647655}
}
Kostenlos starten →