AutoFL: A Bayesian Game Approach for Autonomous Client Participation in Federated Edge Learning
IEEE Transactions on Mobile Computing · 2024-01 · 23 (1) · 194-208
DOI: 10.1109/tmc.2022.3227014 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Mobile Computing, 2024-01. Verfasst von Miao Hu und 6 weiteren Autor:innen. Die Arbeit wurde bisher 29-mal zitiert (Crossref) und nennt 56 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Science and Technology Planning Project of Guangdong Province, Natural Science Foundation of Guangdong Province.
Abstract (Originalsprache)
Given that devices (i.e., clients) participating in federated edge learning (FEL) are autonomous and resource-constrained in nature, it is critical to design effective incentive mechanisms to encourage client participation so as to improve the performance of FEL. In this article, we aim to boost the FEL training efficiency by answering how much compute resource should clients autonomously contribute to maximize their utilities. To this end, we develop AutoFL, an autonomous client participation decision framework for federated learning at the network edge without assuming that each client possesses complete information. We first model the problem of autonomous client participation as a Bayesian game with incomplete information, where each player in the game is associated with a set of types according to network conditions. We optimize an individual client's decision based on the dynamics of the population estimated following the Bayes rule. We prove that AutoFL can converge to a unique Bayesian Nash equilibrium point. Empirical results on three real datasets show that AutoFL achieves a higher model accuracy with only 15.5-24.5% model aggregation time per global training round, and its energy cost saving on mobile devices is 82.2-86.8% compared to the state-of-the-art algorithms. Moreover, we can achieve a 2.75-3.2x long-term fairness compared to classical solutions.
Förderung
- National Natural Science Foundation of China
- Science and Technology Planning Project of Guangdong Province
- Natural Science Foundation of Guangdong Province
- Guangdong Introducing Innovative and Entrepreneurial Teams
- Guangdong Provincial Pearl River Talents Program
- Shenzhen Fundamental Research
Zitieren
APA
Hu, M., Yang, W., Luo, Z., Liu, X., Zhou, Y., Chen, X., & Wu, D. (2024). AutoFL: A Bayesian Game Approach for Autonomous Client Participation in Federated Edge Learning. IEEE Transactions on Mobile Computing, 23(1), 194-208. https://doi.org/10.1109/tmc.2022.3227014
Vancouver
Hu M, Yang W, Luo Z, Liu X, Zhou Y, Chen X, et al. AutoFL: A Bayesian Game Approach for Autonomous Client Participation in Federated Edge Learning. IEEE Transactions on Mobile Computing. 2024;23(1):194-208. doi:10.1109/tmc.2022.3227014
BibTeX
@article{Hu2024,
author = {Hu, Miao and Yang, Wenzhuo and Luo, Zhenxiao and Liu, Xuezheng and Zhou, Yipeng and Chen, Xu and Wu, Di},
title = {AutoFL: A Bayesian Game Approach for Autonomous Client Participation in Federated Edge Learning},
journal = {IEEE Transactions on Mobile Computing},
year = {2024},
volume = {23},
number = {1},
pages = {194--208},
doi = {10.1109/tmc.2022.3227014}
}
EUVIMED