EUVIMEDEUVIMED entdecken
Crossref

FedQClip: Accelerating Federated Learning via Quantized Clipped SGD

Zhihao Qu, Ninghui Jia, Baoliu Ye, Shihong Hu, Song Guo

IEEE Transactions on Computers · 2025-02 · 74 (2) · 717-730

DOI: 10.1109/tc.2024.3477972 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Computers, 2025-02. Verfasst von Zhihao Qu und 4 weiteren Autor:innen. Die Arbeit wurde bisher 21-mal zitiert (Crossref) und nennt 40 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Jiangsu Province of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Federated Learning (FL) has emerged as a promising technique for collaboratively training machine learning models among multiple participants while preserving privacy-sensitive data. However, the conventional parameter server architecture presents challenges in terms of communication overhead when employing iterative optimization methods such as Stochastic Gradient Descent (SGD). Although communication compression techniques can reduce the traffic cost of FL during each training round, they often lead to degraded convergence rates, mainly due to compression errors and data heterogeneity. To address these issues, this paper presents FedQClip, an innovative approach that combines quantization and Clipped SGD. FedQClip leverages an adaptive step size inversely proportional to the $\ell_{2}$ ℓ 2 norm of the gradient, effectively mitigating the negative impacts of quantized errors. Additionally, clipped operations can be applied locally and globally to further expedite training. Theoretical analyses provide evidence that, even under the settings of Non-IID (non-independent and identically distributed) data, FedQClip achieves a convergence rate of $\mathcal{O}(\frac{1}{\sqrt{T}})$ O ( 1 T ) , effectively addressing the convergence degradation caused by compression errors. Furthermore, our theoretical analysis highlights the importance of selecting an appropriate number of local updates to enhance the convergence of FL training. Through extensive experiments, we demonstrate that FedQClip outperforms state-of-the-art methods in terms of communication efficiency and convergence rate.

Förderung

  • National Natural Science Foundation of China
  • Natural Science Foundation of Jiangsu Province of China

Zitieren

APA

Qu, Z., Jia, N., Ye, B., Hu, S., & Guo, S. (2025). FedQClip: Accelerating Federated Learning via Quantized Clipped SGD. IEEE Transactions on Computers, 74(2), 717-730. https://doi.org/10.1109/tc.2024.3477972

Vancouver

Qu Z, Jia N, Ye B, Hu S, Guo S. FedQClip: Accelerating Federated Learning via Quantized Clipped SGD. IEEE Transactions on Computers. 2025;74(2):717-730. doi:10.1109/tc.2024.3477972

BibTeX
@article{Qu2025,
  author = {Qu, Zhihao and Jia, Ninghui and Ye, Baoliu and Hu, Shihong and Guo, Song},
  title = {FedQClip: Accelerating Federated Learning via Quantized Clipped SGD},
  journal = {IEEE Transactions on Computers},
  year = {2025},
  volume = {74},
  number = {2},
  pages = {717--730},
  doi = {10.1109/tc.2024.3477972}
}
Kostenlos starten →