Node Selection Toward Faster Convergence for Federated Learning on Non-IID Data

Abstract

Federated Learning (FL) is a distributed learning paradigm that enables a large number of resource-limited nodes to collaboratively train a model without data sharing. The non-independent-and-identically-distributed (non-i.i.d.) data samples invoke discrepancies between the global and local objectives, making the FL model slow to converge. In this paper, we proposedOptimal Aggregationalgorithm for better aggregation, which finds out the optimal subset of local updates of participating nodes in each global round, by identifying and excluding the adverse local updates via checking the relationship between the local gradient and the global gradient. Then, we proposed aProbabilisticNodeSelection framework (FedPNS) to dynamically change the probability for each node to be selected based on the output ofOptimal Aggregation.FedPNScan preferentially select nodes that propel faster model convergence. The convergence rate improvement ofFedPNSover the commonly adopted Federated Averaging (FedAvg) algorithm is analyzed theoretically. Experimental results demonstrate the effectiveness ofFedPNSin accelerating the FL convergence rate, as compared toFedAvgwith random node selection.