ClusterFL: A Similarity-Aware Federated Learning System for Human Activity Recognition

ClusterFL can efficiently drop out the nodes that converge slower or have little correlation with other nodes in each cluster, significantly speeding up the convergence while maintaining the accuracy performance
Reference
Xiaomin OuyangZhiyuan Xie, Jiayu Zhou, Jianwei Huang, and Guoliang Xing, ClusterFL: A Similarity-Aware Federated Learning System for Human Activity Recognition. In The 19th Annual International Conference on Mobile Systems, Applications, and Services (Acceptance rate: 36/166=21.6%), 2021