通过调控本地梯度动态，提升异构环境下联邦学习的稳定性。
Local Gradient Regulation Stabilizes Federated Learning under Client Heterogeneity
- 从客户端优化角度出发，调节梯度方向以缓解数据异构带来的不稳定
- 在LC25000医疗影像数据集上，显著改善多方法下的收敛性能
- 无需额外通信开销，适合实际部署的分布式医疗或金融场景
联邦学习（FL）可在不共享原始数据的前提下实现跨分布式客户端的协同建模，但其稳定性在真实场景中受到统计异构性的根本挑战。本文发现，客户端异构主要通过扭曲客户端优化过程中的本地梯度动态，引发系统性漂移，并在通信轮次间累积，阻碍全局收敛。这一观察表明，本地梯度是稳定异构联邦学习系统的关键调控变量。基于此，我们提出一种通用的客户端视角，通过无额外通信开销的方式调节本地梯度贡献。受群体智能启发，我们实现该思想为探索-收敛梯度重聚合（ECGR），平衡对齐与非对齐梯度成分，在保留有效更新的同时抑制扰动效应。理论分析与大量实验（包括在LC25000医疗影像数据集上的评估）表明，调节本地梯度动态可一致地稳定多种主流方法在异构数据分布下的联邦学习过程。
原文摘要 · Abstract (English)
Federated learning (FL) enables collaborative model training across distributed clients without sharing raw data, yet its stability is fundamentally challenged by statistical heterogeneity in realistic deployments. Here, we show that client heterogeneity destabilizes FL primarily by distorting local gradient dynamics during client-side optimization, causing systematic drift that accumulates across communication rounds and impedes global convergence. This observation highlights local gradients as a key regulatory lever for stabilizing heterogeneous FL systems. Building on this insight, we develop a general client-side perspective that regulates local gradient contributions without incurring additional communication overhead. Inspired by swarm intelligence, we instantiate this perspective through Exploratory--Convergent Gradient Re-aggregation (ECGR), which balances well-aligned and misaligned gradient components to preserve informative updates while suppressing destabilizing effects. Theoretical analysis and extensive experiments, including evaluations on the LC25000 medical imaging dataset, demonstrate that regulating local gradient dynamics consistently stabilizes federated learning across state-of-the-art methods under heterogeneous data distributions.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。