Personalized Federated Learning under Mixture Distributions

Publication Date: 7/29/2023

Event: The 40th International Conference on Machine Learning (ICML 2023)

Reference: pp. 1-20, 2023

Authors: Yue Wu, University of California, Los Angeles; Shuaicheng Zhang, Virginia Tech; Wenchao Yu, NEC Laboratories America, Inc.; Yanchi Liu, NEC Laboratories America, Inc.; Quanquan Gu, University of California, Los Angeles; Dawei Zhou, Virginia Tech; Haifeng Chen, NEC Laboratories America, Inc.; Wei Cheng, NEC Laboratories America, Inc.

Abstract: The recent trend towards Personalized Federated Learning (PFL) has garnered significant attention as it allows for the training of models that are tailored to each client while maintaining data privacy. However, current PFL techniques primarily focus on modeling the conditional distribution heterogeneity (i.e. concept shift), which can result in suboptimal performance when the distribution of input data across clients diverges (i.e. covariate shift). Additionally, these techniques often lack the ability to adapt to unseen data, further limiting their effectiveness in real-world scenarios. To address these limitations, we propose a novel approach, FedGMM, which utilizes Gaussian mixture models (GMM) to effectively fit the input data distributions across diverse clients. The model parameters are estimated by maximum likelihood estimation utilizing a federated Expectation-Maximization algorithm, which is solved in closed form and does not assume gradient similarity. Furthermore, FedGMM possesses an additional advantage of adapting to new clients with minimal overhead, and it also enables uncertainty quantification. Empirical evaluations on synthetic and benchmark datasets demonstrate the superior performance of our method in both PFL classification and novel sample detection.

Publication Link: