Federated learning is a distributed learning method protecting the data privacy.It trains a global shared model by aggregating local model updates from multiple clients.To deal with the problem of heterogeneity in data distribution among clients, personalized federated learning is an effective way introducing a personalized component into the traditional federated learning framework and enabling each client to train a local model according to its specific needs.In this paper, a new personalized federated learning method combining the stratified sampling strategy is proposed to improve the efficiency of model training and reduce communication complexity.Convergence of the algorithm is established and simulation experiments are implemented to show that reasonable selection of the number of groups and sampling strategy can significantly accelerate the convergence of the algorithm and improve its accuracy.
McMahanH B, MooreE, RamageD, et al.Communication-efficient learning of deep networks from decentralized data [C]// Proceedings of the 20th International Conference on Artificial Intelligence. New York: PMLR, 2017: 1273-1282.
[2]
ElbirA M, ShiW, MishraK V, et al.Federated multi-task learning for THz wideband channel and DoA estimation [C]// 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW).Piscataway: IEEE, 2023: 1-5.
[3]
SahuA K, LiT, ZaheerM, et al.Federated optimization in heterogeneous networks [J].Proc Mach Learn Syst, 2020, 2: 429-450.
[4]
ReddiS, CharlesZ, ZaheerM, et al.Adaptive federated optimization [EB/OL].[2024-09-24].
[5]
WangJ, CharlesZ, XuZ, et al.A field guide to federated optimization [EB/OL].[2024-09-24].
[6]
BonawitzK, EichnerH, GrieskampW, et al.Towards federated learning at scale: System design [J].Proceedings of Machine Learning and Systems, 2019, 1: 374-388.
[7]
McMahanH B, RamageD, TalwarK, et al.Learning differentially private recurrent language models [EB/ OL].[2024-09-24].
[8]
BonawitzK, IvanovV, KreuterB, et al.Practical secure aggregation for federated learning on user-held data [EB/OL].[2024-09-24].
[9]
ChoY J, SharmaP, JoshiG, et al.On the convergence of federated averaging with cyclic client participation [C]// Proceedings of the 40th International Conference on Machine Learning. New York: PMLR, 2023: 5677-5721.
[10]
ZhaoZ, FengC, HongW, et al.Federated learning with non-IID data in wireless networks [J].IEEE Trans Wireless Commun, 2021, 21(3): 1927-1942.
[11]
ZhangZ, MaS, YangZ, et al.Robust semi-supervised federated learning for images automatic recognition in Internet of Drones [J].IEEE Internet of Things Journal, 2022, 10(7): 5733-5746.
FallahA, MokhtariA, OzdaglarA.Personalized federated learning with theoretical guarantees: A model-agnostic meta-learning approach [J].Advances in Neural Information Processing Systems, 2020, 33: 3557-3568.
[14]
ZhangJ, HuaY, WangH, et al.GPFL: Simultaneously learning global and personalized feature information for personalized federated learning [C]// Proceedings of the IEEE/CVF International Conference on Computer Vision.Piscataway: IEEE, 2023: 5041-5051.
[15]
HsuT-M H, QiH, BrownM.Measuring the effects of non-identical data distribution for federated visual classification [EB/OL].[2024-09-24].