FedSkel: Efficient Federated Learning on Heterogeneous Systems with Skeleton Gradients Update

TL;DR

FedSkel introduces a method for federated learning that updates only essential model parts, significantly improving training speed and reducing communication costs on heterogeneous edge devices while maintaining accuracy.

Contribution

This work presents FedSkel, a novel skeleton gradient update approach that enhances efficiency in federated learning on diverse edge systems.

Findings

Achieves up to 5.52× speedup in CONV layer back-propagation.

Reduces overall training time by 1.82×.

Cuts communication costs by 64.8%.

Abstract

Federated learning aims to protect users' privacy while performing data analysis from different participants. However, it is challenging to guarantee the training efficiency on heterogeneous systems due to the various computational capabilities and communication bottlenecks. In this work, we propose FedSkel to enable computation-efficient and communication-efficient federated learning on edge devices by only updating the model's essential parts, named skeleton networks. FedSkel is evaluated on real edge devices with imbalanced datasets. Experimental results show that it could achieve up to 5.52$\times$ speedups for CONV layers' back-propagation, 1.82$\times$ speedups for the whole training process, and reduce 64.8% communication cost, with negligible accuracy loss.