FedGroup: Accurate Federated Learning via Decomposed Similarity-Based Clustering

Federated Learning (FL) enables the multiple participating devices to collaboratively contribute to a global neural network model while keeping the training data locally. Unlike the centralized training setting, the non-IID and imbalanced (statistical heterogeneity) training data of FL is distributed in the federated network, which will increase the divergences between the local models and global model, further degrading performance. In this paper, we propose a novel clustered federated learning (CFL) framework FedGroup based on a similarity-based client clustering strategy, in which we 1) group the training of clients based on the similarities between the clients' optimize directions for high training performance; 2) reduce the complexity of client clustering algorithm by decomposing the high-dimension low-sample size (HDLSS) direction vectors. 3) implement a newcomer device cold start mechanism based on the auxiliary global model for framework scalability and practicality. FedGroup can achieve improvements by dividing joint optimization into groups of sub-optimization, and can be combined with FedProx, the state-of-the-art federated optimization algorithm. We evaluate FedGroup and FedGrouProx (combined with FedProx) on several open datasets. The experimental results show that our proposed frameworks significantly improving absolute test accuracy by +14.7% on FEMNIST compared to FedAvg, +5.4% on Sentiment140 compared to FedProx.