TinyFEL: Communication, Computation, and Memory Efficient Tiny Federated Edge Learning via Model Sparse Update
IEEE Internet of Things Journal · 01.04.2025 · 12 (7) · 8247-8260
DOI: 10.1109/jiot.2024.3499375 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Internet of Things Journal, erschienen am 01.04.2025. Verfasst von Qimei Chen und 5 weiteren Autor:innen. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 38 Literaturangaben. Gefördert von: Key Research and Development Plan of Hubei Province, Wuhan Science and Technology Achievement Transformation Project, Digital Information Infrastructure Operation and Maintenance Innovation Base Co-raised by Wuhan University, Fiberhome Telecommunication Technologies Company, L….
Abstract (Originalsprache)
Federated edge learning (FEL) is regarded as a promising distributed machine learning paradigm to reduce transmission latency and resources as well as preserve raw data privacy by collaboratively training local deep learning models across multiple edge devices. However, with the development of artificial intelligence (AI) technologies, the size of neural network models grows exponentially with their parameters to meet variable application requirements, which poses significant challenges to the computation, communication, and memory abilities of edge devices. Existing designs typically focus on either communication or computation efficiency without caring each device’s memory ability. To deal with the above issues, we first introduce a novel model sparse update enabled tiny FEL (TinyFEL) architecture, which terminates the backpropagation early in local model training processes. Therefore, the proposed TinyFEL can reduce local memory occupation and lessen the communication-and-computation burden. Furthermore, we propose a parameter splitting mechanism instead of transmitting the full model, only a part of updated layers of parameters is transmitted for aggregation, which significantly reduced the communication overheads. Thereafter, we develop a communication and computation latency minimization problem to accelerate the training of TinyFEL. To this end, we theoretically analyze the convergence performance of TinyFEL, which unveils the mathematical relationship among sparse update ratio assignment, device selection, and learning performance. Then, a joint sparse update ratio assignment, device selection, and resource allocation strategy is introduced based on the alternating direction method of multipliers (ADMMs) and block coordinate descent (BCD) algorithms. Numerical results indicate that our proposed TinyFEL can reduce training memory occupation by over 40% than the traditional FEL at the cost of negligible accuracy loss.
Förderung
- Key Research and Development Plan of Hubei Province
- Wuhan Science and Technology Achievement Transformation Project
- Digital Information Infrastructure Operation and Maintenance Innovation Base Co-raised by Wuhan University, Fiberhome Telecommunication Technologies Company, L…
- National Natural Science Foundation of China
- Guangdong Major Project of Basic and Applied Basic Research
- Basic and Applied Basic Research Foundation of Guangdong Province
- Shenzhen–Hong Kong–Macau Technology Research Programme
- Longgang District Special Funds for Science and Technology Innovation
Zitieren
APA
Chen, Q., Cheng, H., Liang, Y., Zhu, G., Li, M., & Jiang, H. (2025). TinyFEL: Communication, Computation, and Memory Efficient Tiny Federated Edge Learning via Model Sparse Update. IEEE Internet of Things Journal, 12(7), 8247-8260. https://doi.org/10.1109/jiot.2024.3499375
Vancouver
Chen Q, Cheng H, Liang Y, Zhu G, Li M, Jiang H. TinyFEL: Communication, Computation, and Memory Efficient Tiny Federated Edge Learning via Model Sparse Update. IEEE Internet of Things Journal. 2025;12(7):8247-8260. doi:10.1109/jiot.2024.3499375
BibTeX
@article{Chen2025,
author = {Chen, Qimei and Cheng, Han and Liang, Yipeng and Zhu, Guangxu and Li, Ming and Jiang, Hao},
title = {TinyFEL: Communication, Computation, and Memory Efficient Tiny Federated Edge Learning via Model Sparse Update},
journal = {IEEE Internet of Things Journal},
year = {2025},
volume = {12},
number = {7},
pages = {8247--8260},
doi = {10.1109/jiot.2024.3499375}
}
EUVIMED