Sparse-FCL: Sparse Federated Continual Learning for Evolving Mobile Edge Computing Environments

Yan Liu

doi:10.1109/TSC.2025.3583174

2025-07-01

PRE

DOI：10.1109/TSC.2025.3583174

原文链接

原文求助

摘要

En 中文

移动边缘计算（MEC）利用嵌入式设备提供低延迟、低成本高效的智能服务，例如自动驾驶。联邦学习允许MEC设备在不暴露私有数据的情况下协作学习全局模型。然而，MEC设备运行的实际动态环境中可能导致先前学习知识的灾难性遗忘。尽管持续学习已被应用于联邦学习中以保留过去数据的知识，但在将其扩展应用于演变的MEC环境时，由于资源受限设备的难以承受的通信和计算成本，存在显著障碍。我们通过提出Sparse-FCL解决了这一问题，这是一种联邦持续学习框架，利用稀疏训练来减少训练开销并提高模型性能。具体而言，为了保留特定任务的一般化知识，我们引入了一个通过多设备协作模块的渐进神经元选择方法，该方法在每个持续学习任务中逐步选择在多个设备上显示出重要性的关键神经元。此外，我们提出了一个任务自适应拓扑探索模块，旨在为联邦持续学习场景提供更灵活的动态稀疏训练配置。在FCL基准测试上的实验表明，Sparse-FCL在高稀疏度水平下具有优异的准确性，并且与现有的联邦持续学习方法相比，其在通信、计算和存储开销方面分别减少了80.4%、89.5%和19.7%。

Keyword:

Federated learning

continual learning

sparse training

mobile edge computing

期刊

IEEE Transactions on Services Computing

IF:

5.8

论文数:

2.2K

被引数:

6.5K

机构

Harbin Engineering University

学者数:

1.9W

论文数: 1.3W

被引数: 1.3W

Northwestern Polytechnical University

学者数:

4.6W

论文数: 3.7W

被引数: 5.3W

引用论文

WAVE: Edge-Device Cooperated Real-Time Object Detection for Open-Air Applications WAVE: 面向露天应用的边缘设备协作实时目标检测

IEEE TRANSACTIONS ON MOBILE COMPUTING

IF9.2

2023-07-01

PREAI

Dong, Liang; Yang, Zheng; Cai, Xinjun; Zhao, Yi; Ma, Qiang; Miao, Xin

The stability-plasticity dilemma: investigating the continuum from catastrophic forgetting to age-limited learning effects

Frontiers in Psychology

IF0

2013-01-01

OAAI

Martial Mermillod; Aurélia Bugaiska; Patrick Bonin

HARKE: Human Activity Recognition from Kinetic Energy Harvesting Data in Wearable Devices HARKE: 从可穿戴设备的动能收集数据中识别人体活动

IEEE TRANSACTIONS ON MOBILE COMPUTING

IF9.2

2018-06-01

100

PREAI

Khalifa, Sara; Lan, Guohao; Hassan, Mahbub; Seneviratne, Aruna; Das, Sajal K.

Wiley Series in Probability and Statistics 威利概率与统计系列

IF0

2025-01-25

PREAI

Diffusion markers of dendritic density and arborization in gray matter predict differences in intelligence

NATURE COMMUNICATIONS

IF15.7

2018-05-15

101

OAAI

Genc, Erhan; Fraenz, Christoph; Schlueter, Caroline; Friedrich, Patrick; Hossiep, Ruediger; Voelkle, Manuel C.; Ling, Josef M.; Guentuerkuen, Onur; Jung, Rex E.

FedASMU: Efficient Asynchronous Federated Learning with Dynamic Staleness-Aware Model Update

Proceedings of the AAAI Conference on Artificial Intelligence

IF0

2024-03-24

OAAI

Ji Liu; Juncheng Jia; Tianshi Che; Chao Huo; Jiaxiang Ren; Yang Zhou; Huaiyu Dai; Dejing Dou

Advances and Open Problems in Federated Learning

IF0

2021-01-01

OAAI

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

Non-IID data and Continual Learning processes in Federated Learning: A long road ahead 联合学习中的非IID数据和持续学习过程: 漫长的道路

INFORMATION FUSION

IF15.5

2022-12-01

OAAI

Criado, Marcos F.; Casado, Fernando E.; Iglesias, Roberto; V. Regueiro, Carlos; Barro, Senen

PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruning PackNet: 通过迭代修剪将多个任务添加到单个网络

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition

IF0

2018-06-01

OAAI

Arun Mallya; Svetlana Lazebnik