FedSTaS: Client Stratification and Client Level Sampling for Efficient Federated Learning

Federated learning (FL) is a machine learning methodology that involves the collaborative training of a global model across multiple decentralized clients in a privacy-preserving way. Several FL methods are introduced to tackle communication inefficiencies but do not address how to sample participating clients in each round effectively and in a privacy-preserving manner. In this paper, we propose $\mathit{\text{FedSTaS}}$, a client and data-level sampling method inspired by $\mathit{\text{FedSTS}}$ and $\mathit{\text{FedSampling}}$. In each federated learning round, $\mathit{\text{FedSTaS}}$ stratifies clients based on their compressed gradients, re-allocate the number of clients to sample using an optimal Neyman allocation, and sample local data from each participating clients using a data uniform sampling strategy. Experiments on three datasets show that $\mathit{\text{FedSTaS}}$ can achieve higher accuracy scores than those of $\mathit{\text{FedSTS}}$ within a fixed number of training rounds.