A Unified Solution to Diverse Heterogeneities in One-Shot Federated Learning
Files
(Published version)
Date
2025
Authors
Bai, J.
Song, Y.
Wu, D.
Sajjanhar, A.
Xiang, Y.
Zhou, W.
Tao, X.
Li, Y.
Li, Y.
Editors
Advisors
Journal Title
Journal ISSN
Volume Title
Type:
Conference paper
Citation
Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2 (KDD '25), 2025, vol.2, pp.71-82
Statement of Responsibility
Jun Bai, Yiliao Song, Di Wu, Atul Sajjanhar, Yong Xiang, Wei Zhou, Xiaohui Tao, Yan Li, Yue Li
Conference Name
ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) (3 Aug 2025 - 4 Aug 2025 : Toronto, ON, Canada)
Abstract
One-Shot Federated Learning (OSFL) restricts communication between the server and clients to a single round, significantly reducing communication costs and minimizing privacy leakage risks compared to traditional Federated Learning (FL), which requires multiple rounds of communication. However, existing OSFL frameworks remain vulnerable to distributional heterogeneity, as they primarily focus on model heterogeneity while neglecting data heterogeneity. To bridge this gap, we propose FedHydra, a unified, data-free, OSFL framework designed to effectively address both model and data heterogeneity. Unlike existing OSFL approaches, FedHydra introduces a novel two-stage learning mechanism. Specifically, it incorporates model stratification and heterogeneity-aware stratified aggregation to mitigate the challenges posed by both model and data heterogeneity. By this design, the data and model heterogeneity issues are simultaneously monitored from different aspects during learning. Consequently, FedHydra can effectively mitigate both issues by minimizing their inherent conflicts. We compared FedHydra with five SOTA baselines on four benchmark datasets. Experimental results show that our method outperforms the previous OSFL methods in both homogeneous and heterogeneous settings. The code is available at https://github.com/Jun-B0518/FedHydra.
School/Discipline
Dissertation Note
Provenance
Description
Access Status
Rights
© 2025 Copyright held by the owner/author(s). This work is licensed under a Creative Commons Attribution 4.0 International License.