EUVIMEDEUVIMED entdecken
Crossref

AdpFL: A Privacy-Preserving Federated Learning Framework Through Adaptive Model Pruning on Non-IID Data

Dewei Ning, Yong-Feng Ge, Elisa Bertino, Zhonglong Zheng, Yunliang Jiang, Hua Wang

IEEE Transactions on Services Computing · 2026-09 · 19 (5) · 3269-3282

DOI: 10.1109/tsc.2026.3686986 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Services Computing, 2026-09. Verfasst von Dewei Ning und 5 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 65 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Federated learning (FL) has shown great potential, especially with the rise of complex foundation models and growing privacy needs. FL has experienced challenges, including high communication costs, privacy concerns of user data, and the complexities of non-independent and identically distributed (non-IID) data. This paper aims to address these challenges through our proposed FL framework called Adaptive Pruning FL (AdpFL). AdpFL conducts an exploration phase prior to the formal FL process, during which each local dataset’s deviation and corresponding threshold are computed locally. These results are then utilised to perform model pruning during the formal FL training. By doing so, the method achieves full adaptivity while simultaneously addressing challenges related to communications costs and non-IID data. Meanwhile, AdpFL enhances pruning by replacing part of the parameters with zeros, thereby preventing membership inference attacks (MIA) and ensuring privacy in both local and global models. The extensive experiments on various datasets and FL tasks demonstrate that our method not only efficiently accelerates global convergence but also achieves superior accuracy and robustness compared to traditional methods. AdpFL reduces computational overhead and decreases the effective update size (i.e., the proportion of non-zero parameters) by approximately half while improving global accuracy by up to 20%, and it also lowers the accuracy of white-box MIA attacks across various models. These improvements make FL more efficient, scalable, and practical in real-world situations.

Zitieren

APA

Ning, D., Ge, Y., Bertino, E., Zheng, Z., Jiang, Y., & Wang, H. (2026). AdpFL: A Privacy-Preserving Federated Learning Framework Through Adaptive Model Pruning on Non-IID Data. IEEE Transactions on Services Computing, 19(5), 3269-3282. https://doi.org/10.1109/tsc.2026.3686986

Vancouver

Ning D, Ge Y, Bertino E, Zheng Z, Jiang Y, Wang H. AdpFL: A Privacy-Preserving Federated Learning Framework Through Adaptive Model Pruning on Non-IID Data. IEEE Transactions on Services Computing. 2026;19(5):3269-3282. doi:10.1109/tsc.2026.3686986

BibTeX
@article{Ning2026,
  author = {Ning, Dewei and Ge, Yong-Feng and Bertino, Elisa and Zheng, Zhonglong and Jiang, Yunliang and Wang, Hua},
  title = {AdpFL: A Privacy-Preserving Federated Learning Framework Through Adaptive Model Pruning on Non-IID Data},
  journal = {IEEE Transactions on Services Computing},
  year = {2026},
  volume = {19},
  number = {5},
  pages = {3269--3282},
  doi = {10.1109/tsc.2026.3686986}
}
Kostenlos starten →