WAFFLE: Weighted Averaging for Personalized Federated Learning
This work addresses personalized federated learning for clients with non-identical data distributions, presenting an incremental improvement over existing methods.
The paper tackles the problem of data heterogeneity in federated learning by introducing WAFFLE, a personalized algorithm that uses weighted averaging based on Euclidean distances between client updates, achieving improved or comparable accuracy with faster convergence on MNIST and CIFAR10 datasets under concept shift and label skew.
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.