Federated Split Learning with Model Pruning and Gradient Quantization in Wireless Networks

TL;DR

This paper introduces a lightweight federated split learning scheme that reduces resource consumption on edge devices through model pruning, gradient quantization, and dropout, while maintaining effective collaborative training in wireless networks.

Contribution

It proposes a novel lightweight FedSL approach combining model pruning, gradient quantization, and dropout to improve efficiency in resource-constrained wireless edge devices.

Findings

The scheme reduces computation and communication overhead.

Theoretical analysis confirms convergence performance.

Simulations demonstrate improved efficiency and effectiveness.

Abstract

As a paradigm of distributed machine learning, federated learning typically requires all edge devices to train a complete model locally. However, with the increasing scale of artificial intelligence models, the limited resources on edge devices often become a bottleneck for efficient fine-tuning. To address this challenge, federated split learning (FedSL) implements collaborative training across the edge devices and the server through model splitting. In this paper, we propose a lightweight FedSL scheme, that further alleviates the training burden on resource-constrained edge devices by pruning the client-side model dynamicly and using quantized gradient updates to reduce computation overhead. Additionally, we apply random dropout to the activation values at the split layer to reduce communication overhead. We conduct theoretical analysis to quantify the convergence performance of the proposed scheme. Finally, simulation results verify the effectiveness and advantages of the proposed lightweight FedSL in wireless network environments.