FedVQCS: Federated Learning via Vector Quantized Compressed Sensing

TL;DR

FedVQCS introduces a communication-efficient federated learning framework that combines vector quantized compressed sensing with sparse signal recovery to significantly reduce communication costs while maintaining high accuracy.

Contribution

It proposes a novel FL framework that integrates dimensionality reduction and vector quantization with optimized parameters for minimal reconstruction error.

Findings

Achieves over 2.4% higher accuracy than state-of-the-art FL methods at low communication cost.

Effectively reduces communication overhead through vector quantization and compressed sensing techniques.

Demonstrates robustness and efficiency on MNIST and FEMNIST datasets.

Abstract

In this paper, a new communication-efficient federated learning (FL) framework is proposed, inspired by vector quantized compressed sensing. The basic strategy of the proposed framework is to compress the local model update at each device by applying dimensionality reduction followed by vector quantization. Subsequently, the global model update is reconstructed at a parameter server by applying a sparse signal recovery algorithm to the aggregation of the compressed local model updates. By harnessing the benefits of both dimensionality reduction and vector quantization, the proposed framework effectively reduces the communication overhead of local update transmissions. Both the design of the vector quantizer and the key parameters for the compression are optimized so as to minimize the reconstruction error of the global model update under the constraint of wireless link capacity. By considering the reconstruction error, the convergence rate of the proposed framework is also analyzed for a non-convex loss function. Simulation results on the MNIST and FEMNIST datasets demonstrate that the proposed framework provides more than a 2.4% increase in classification accuracy compared to state-of-the-art FL frameworks when the communication overhead of the local model update transmission is 0.1 bit per local model entry.