FedLGA: Towards System-Heterogeneity of Federated Learning via Local Gradient Approximation

TL;DR

FedLGA introduces a novel federated learning algorithm that effectively addresses system heterogeneity by approximating local gradients, ensuring robust convergence and outperforming existing methods in diverse device and data settings.

Contribution

This work formalizes the system-heterogeneous FL problem and proposes FedLGA, a gradient approximation-based algorithm with theoretical convergence guarantees under heterogeneity.

Findings

FedLGA achieves better convergence rates than existing methods.

Experimental results show FedLGA outperforms baselines on multiple datasets.

FedLGA effectively handles device and data heterogeneity in federated learning.

Abstract

Federated Learning (FL) is a decentralized machine learning architecture, which leverages a large number of remote devices to learn a joint model with distributed training data. However, the system-heterogeneity is one major challenge in a FL network to achieve robust distributed learning performance, which comes from two aspects: i) device-heterogeneity due to the diverse computational capacity among devices; ii) data-heterogeneity due to the non-identically distributed data across the network. Prior studies addressing the heterogeneous FL issue, e.g., FedProx, lack formalization and it remains an open problem. This work first formalizes the system-heterogeneous FL problem and proposes a new algorithm, called FedLGA, to address this problem by bridging the divergence of local model updates via gradient approximation. To achieve this, FedLGA provides an alternated Hessian estimation method, which only requires extra linear complexity on the aggregator. Theoretically, we show that with a device-heterogeneous ratio $\rho$, FedLGA achieves convergence rates on non-i.i.d. distributed FL training data for the non-convex optimization problems with $\mathcal{O} \left( \frac{(1+\rho)}{\sqrt{ENT}} + \frac{1}{T} \right)$ and $\mathcal{O} \left( \frac{(1+\rho)\sqrt{E}}{\sqrt{TK}} + \frac{1}{T} \right)$ for full and partial device participation respectively, where $E$ is the number of local learning epoch, $T$ is the number of total communication round, $N$ is the total device number and $K$ is the number of selected device in one communication round under partially participation scheme. The results of comprehensive experiments on multiple datasets show that FedLGA outperforms current FL methods against the system-heterogeneity.