Adaptive aggregation for federated learning using representation ability based on feature alignment

doi:10.1016/j.knosys.2025.113560

2025-06-01

PRE

DOI：10.1016/j.knosys.2025.113560

原文链接

原文求助

摘要

En 中文

联邦学习已被成功应用于物联网（IoT）场景。然而，物联网设备间的数据通常是Non-IID（独立同分布），这种数据异质性会严重影响全局模型的收敛速度，甚至导致模型精度显著下降。为解决此问题，本文提出了一种基于特征对齐表示能力的联邦自适应聚合算法。在联邦训练之前，服务器使用Xavier均匀初始化的全局模型计算初始特征锚点，从而使其更适应数据特征。在客户端训练期间，特征对齐方法通过增强特征锚点在统一的特征空间中更新所有客户端模型，从而减少客户端间的数据异质性影响。在服务器联邦聚合过程中引入了与表示能力相关的权重缩放因子，以在存在数据异质性时自适应地聚合本地模型参数。在流行的图像数据集MNIST、Fashion-MNIST、CIFAR-10和CIFAR100上进行了大量实验。实验结果表明，该方法显著提高了全局模型的收敛速度和模型性能，同时有效缓解了数据异质性的不利影响。

Keyword:

Federated aggregation

Feature alignment

Representation ability

Data heterogeneity

期刊

Knowledge-Based Systems

IF:

7.6

论文数:

1.3W

被引数:

4.5W

机构

Beijing Univ Technol

学者数:

2.6K

论文数: 1.2K

被引数: 354

引用论文

Federated learning‐based IoT: A systematic literature review

International Journal of Communication Systems

IF0

2022-04-24

PREAI

Mehdi Hosseinzadeh; Atefeh Hemmati; Amir Masoud Rahmani

A Federated Learning Framework for Fingerprinting-Based Indoor Localization in Multibuilding and Multifloor Environments

IEEE INTERNET OF THINGS JOURNAL

IF8.9

2023-02-01

PREAI

Gao, Bo; Yang, Fan; Cui, Nan; Xiong, Ke; Lu, Yang; Wang, Yuwei

Preserving Data Privacy via Federated Learning: Challenges and Solutions 通过联合学习保护数据隐私: 挑战和解决方案

IEEE CONSUMER ELECTRONICS MAGAZINE

IF4.1

2020-05-01

119

PREAI

Li, Zengpeng; Sharma, Vishal; Mohanty, Saraju P.

Privacy-Preserving Traffic Flow Prediction: A Federated Learning Approach 保护隐私的交通流预测: 一种联合学习方法

IEEE INTERNET OF THINGS JOURNAL

IF8.9

2020-08-01

357

OAAI

Liu, Yi; Yu, James J. Q.; Kang, Jiawen; Niyato, Dusit; Zhang, Shuyu

FedProc: Prototypical contrastive federated learning on non-IID data

FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE

IF6.1

2023-06-01

OAAI

Mu, Xutong; Shen, Yulong; Cheng, Ke; Geng, Xueli; Fu, Jiaxuan; Zhang, Tao; Zhang, Zhiwei

FedFA: Federated Learning With Feature Anchors to Align Features and Classifiers for Heterogeneous Data FedFA: 使用特征锚进行联合学习，以对齐异构数据的特征和分类器

IEEE TRANSACTIONS ON MOBILE COMPUTING

IF9.2

2024-06-01

OAAI

Zhou, Tailin; Zhang, Jun; Tsang, Danny H. K.

Communication-Efficient Federated Learning for Wireless Edge Intelligence in IoT 物联网中无线边缘智能的高效通信联合学习

IEEE INTERNET OF THINGS JOURNAL

IF8.9

2020-07-01

302

PREAI

Mills, Jed; Hu, Jia; Min, Geyong

Model aggregation techniques in federated learning: A comprehensive survey 联合学习中的模型聚合技术: 一项综合调查

FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE

IF6.1

2024-01-01

OAAI

Qi, Pian; Chiaro, Diletta; Guzzo, Antonella; Ianni, Michele; Fortino, Giancarlo; Piccialli, Francesco

A Systematic Literature Review on Federated Machine Learning: From a Software Engineering Perspective 关于联邦机器学习的系统文献综述: 从软件工程的角度

ACM COMPUTING SURVEYS

IF28

2021-05-25

OAAI

Lo, Sin Kit; Lu, Qinghua; Wang, Chen; Paik, Hye-Young; Zhu, Liming

Advances and Open Problems in Federated Learning 联邦学习的进展和开放问题

FOUNDATIONS AND TRENDS IN MACHINE LEARNING

IF25.4

2021-01-01

2.0K

PREAI

Kairouz, Peter; McMahan, H. Brendan; Avent, Brendan; Bellet, Aurelien; Bennis, Mehdi; Bhagoji, Arjun Nitin; Bonawitz, Kallista; Charles, Zachary; Cormode, Graham; Cummings, Rachel; D'Oliveira, Rafael G. L.; Eichner, Hubert; El Rouayheb, Salim; Evans, David; Gardner, Josh; Garrett, Zachary; Gascon, Adria; Ghazi, Badih; Gibbons, Phillip B.; Gruteser, Marco; Harchaoui, Zaid; He, Chaoyang; He, Lie; Huo, Zhouyuan; Hutchinson, Ben; Hsu, Justin; Jaggi, Martin; Javidi, Tara; Joshi, Gauri; Khodak, Mikhail; Konecny, Jakub; Korolova, Aleksandra; Koushanfar, Farinaz; Koyejo, Sanmi; Lepoint, Tancrede; Liu, Yang; Mittal, Prateek; Mohri, Mehryar; Nock, Richard; Ozgur, Ayfer; Pagh, Rasmus; Qi, Hang; Ramage, Daniel; Raskar, Ramesh; Raykova, Mariana; Song, Dawn; Song, Weikang; Stich, Sebastian U.; Sun, Ziteng; Suresh, Ananda Theertha; Tramer, Florian; Vepakomma, Praneeth; Wang, Jianyu; Xiong, Li; Xu, Zheng; Yang, Qiang; Yu, Felix X.; Yu, Han; Zhao, Sen

查看更多内容