Federated Split Learning with Model Pruning and Gradient Quantization in Wireless Networks
This work addresses efficiency bottlenecks for edge devices in distributed machine learning, representing an incremental improvement to federated split learning.
The paper tackles the challenge of resource-constrained edge devices in federated split learning by proposing a lightweight scheme that dynamically prunes client-side models and uses quantized gradients, reducing computation and communication overhead, with simulation results verifying its effectiveness in wireless networks.
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.