arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.18384cs.LGcs.AI
收藏

FedZMG: 在联邦学习中的高效客户端优化

FedZMG: Efficient Client-Side Optimization in Federated Learning

TelSiP Research Laboratory, Department of Electrical and Electronic Engineering, School of Engineering, University of West Attica(TelSiP研究实验室,电气与电子工程系,工程学院,西阿提卡大学)

Fotios Zantalis, Evangelos Zervas, Grigorios Koulouras

AI总结:

FedZMG是一种参数无关的客户端优化算法,通过结构化正则化解决联邦学习中的客户端漂移问题,提升收敛速度和模型性能。

摘要:

联邦学习(FL)允许在边缘设备上进行分布式模型训练,同时保护数据隐私。然而,客户端往往具有非独立同分布(non-IID)的数据,这通常导致客户端漂移,从而降低收敛速度和模型性能。尽管已经提出了自适应优化器来缓解这些影响,但它们通常会引入计算复杂性或通信开销,这在资源受限的物联网环境中是不适用的。本文介绍了联邦零均值梯度(FedZMG),一种新的、参数无关的客户端优化算法,旨在通过在优化空间中进行结构化正则化来解决客户端漂移问题。在梯度集中化思想的基础上,FedZMG将局部梯度投影到零均值超平面,有效中和了异构数据分布中固有的“强度”或“偏差”偏移,而无需额外的通信或超参数调整。提供了理论分析,证明FedZMG减少了有效梯度方差,并保证了比标准FedAvg更紧的收敛界限。在EMNIST、CIFAR100和Shakespeare数据集上的广泛实证评估表明,FedZMG在收敛速度和最终验证准确率方面优于基线FedAvg和自适应优化器FedAdam,特别是在高度非IID设置中。

英文摘要:

Federated Learning (FL) enables distributed model training on edge devices while preserving data privacy. However, clients tend to have non-Independent and Identically Distributed (non-IID) data, which often leads to client-drift, and therefore diminishing convergence speed and model performance. While adaptive optimizers have been proposed to mitigate these effects, they frequently introduce computational complexity or communication overhead unsuitable for resource-constrained IoT environments. This paper introduces Federated Zero Mean Gradients (FedZMG), a novel, parameter-free, client-side optimization algorithm designed to tackle client-drift by structurally regularizing the optimization space. Advancing the idea of Gradient Centralization, FedZMG projects local gradients onto a zero-mean hyperplane, effectively neutralizing the "intensity" or "bias" shifts inherent in heterogeneous data distributions without requiring additional communication or hyperparameter tuning. A theoretical analysis is provided, proving that FedZMG reduces the effective gradient variance and guarantees tighter convergence bounds compared to standard FedAvg. Extensive empirical evaluations on EMNIST, CIFAR100, and Shakespeare datasets demonstrate that FedZMG achieves better convergence speed and final validation accuracy compared to the baseline FedAvg and the adaptive optimizer FedAdam, particularly in highly non-IID settings.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑