Figure - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Download
View publication
Tuned hyper-parameters in the CIFAR-10 experiment for the global cluster models, the local models and the gating model.

Tuned hyper-parameters in the CIFAR-10 experiment for the global cluster models, the local models and the gating model.

Source publication
Figure 1: Our approach adjusts to non-Independent and Identically...
Tuned hyper-parameters in the CIFAR-10 experiment for the global...
Adaptive Expert Models for Personalization in Federated Learning
Preprint
Full-text available
  • Jun 2022
Federated Learning (FL) is a promising framework for distributed learning when data is private and sensitive. However, the state-of-the-art solutions in this framework are not optimal when data is heterogeneous and non-Independent and Identically Distributed (non-IID). We propose a practical and robust approach to personalization in FL that adjusts...
Cite
Download full-text

Context in source publication

Context 1
... show that our method is still robust in the fully non-IID case when p = 1. See Table 1 for tuned hyperparameters in the CIFAR-10 experiment. ...

Similar publications

Figure 2: Comparison results of the convergence rate of different...
Results on MNIST, FMNIST and CIFAR-10. Best results are bolded.
Results of pFedBayes on Medium dataset (MNIST).
Comparison algorithm results on the medium dataset (MNIST).
Personalized Federated Learning via Variational Bayesian Inference
Preprint
Full-text available
  • Jun 2022
Federated learning faces huge challenges from model overfitting due to the lack of data and statistical diversity among clients. To address these challenges, this paper proposes a novel personalized federated learning method via Bayesian variational inference named pFedBayes. To alleviate the overfitting, weight uncertainty is introduced to neural...
View