GC-Fed: Gradient Centralized Federated Learning with Partial Client Participation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Seo, Jungwon, Catak, Ferhat Ozgur, Rong, Chunming, Hong, Kibeom, Kim, Minhoe
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911438835023872
author Seo, Jungwon
Catak, Ferhat Ozgur
Rong, Chunming
Hong, Kibeom
Kim, Minhoe
author_facet Seo, Jungwon
Catak, Ferhat Ozgur
Rong, Chunming
Hong, Kibeom
Kim, Minhoe
contents Federated Learning (FL) enables privacy-preserving multi-source information fusion (MSIF) but is challenged by client drift in highly heterogeneous data settings. Many existing drift-mitigation strategies rely on reference-based techniques--such as gradient adjustments or proximal loss--that use historical snapshots (e.g., past gradients or previous global models) as reference points. When only a subset of clients participates in each training round, these historical references may not accurately capture the overall data distribution, leading to unstable training. In contrast, our proposed Gradient Centralized Federated Learning (GC-Fed) employs a hyperplane as a historically independent reference point to guide local training and enhance inter-client alignment. GC-Fed comprises two complementary components: Local GC, which centralizes gradients during local training, and Global GC, which centralizes updates during server aggregation. In our hybrid design, Local GC is applied to feature-extraction layers to harmonize client contributions, while Global GC refines classifier layers to stabilize round-wise performance. Theoretical analysis and extensive experiments on benchmark FL tasks demonstrate that GC-Fed effectively mitigates client drift and achieves up to a 20% improvement in accuracy under heterogeneous and partial participation conditions.
format Preprint
id arxiv_https___arxiv_org_abs_2503_13180
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle GC-Fed: Gradient Centralized Federated Learning with Partial Client Participation
Seo, Jungwon
Catak, Ferhat Ozgur
Rong, Chunming
Hong, Kibeom
Kim, Minhoe
Machine Learning
Artificial Intelligence
Distributed, Parallel, and Cluster Computing
Federated Learning (FL) enables privacy-preserving multi-source information fusion (MSIF) but is challenged by client drift in highly heterogeneous data settings. Many existing drift-mitigation strategies rely on reference-based techniques--such as gradient adjustments or proximal loss--that use historical snapshots (e.g., past gradients or previous global models) as reference points. When only a subset of clients participates in each training round, these historical references may not accurately capture the overall data distribution, leading to unstable training. In contrast, our proposed Gradient Centralized Federated Learning (GC-Fed) employs a hyperplane as a historically independent reference point to guide local training and enhance inter-client alignment. GC-Fed comprises two complementary components: Local GC, which centralizes gradients during local training, and Global GC, which centralizes updates during server aggregation. In our hybrid design, Local GC is applied to feature-extraction layers to harmonize client contributions, while Global GC refines classifier layers to stabilize round-wise performance. Theoretical analysis and extensive experiments on benchmark FL tasks demonstrate that GC-Fed effectively mitigates client drift and achieves up to a 20% improvement in accuracy under heterogeneous and partial participation conditions.
title GC-Fed: Gradient Centralized Federated Learning with Partial Client Participation
topic Machine Learning
Artificial Intelligence
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2503.13180