WAFFLE: Weighted Averaging for Personalized Federated Learning

TL;DR

WAFFLE is a novel personalized federated learning algorithm that uses weighted averaging based on client update distances to improve convergence speed and accuracy across heterogeneous data distributions.

Contribution

Introduces WAFFLE, a personalized federated learning method leveraging stochastic control variates and weighted averaging for faster convergence and better personalization.

Findings

WAFFLE outperforms recent personalized FL methods in accuracy.

WAFFLE achieves faster convergence than baseline methods.

Effective across concept shift and label skew data distributions.

Abstract

In federated learning, model personalization can be a very effective strategy to deal with heterogeneous training data across clients. We introduce WAFFLE (Weighted Averaging For Federated LEarning), a personalized collaborative machine learning algorithm that leverages stochastic control variates for faster convergence. WAFFLE uses the Euclidean distance between clients' updates to weigh their individual contributions and thus minimize the personalized model loss on the specific agent of interest. Through a series of experiments, we compare our new approach to two recent personalized federated learning methods--Weight Erosion and APFL--as well as two general FL methods--Federated Averaging and SCAFFOLD. Performance is evaluated using two categories of non-identical client data distributions--concept shift and label skew--on two image data sets (MNIST and CIFAR10). Our experiments demonstrate the comparative effectiveness of WAFFLE, as it achieves or improves accuracy with faster convergence.