Attention-based Feature Compression for CNN Inference Offloading in Edge Computing

TL;DR

This paper introduces AECNN, an autoencoder-based CNN architecture that effectively compresses intermediate features for CNN inference offloading in edge computing, significantly reducing data size while maintaining accuracy.

Contribution

The paper proposes a novel feature compression method using channel attention and entropy encoding, improving offloading efficiency and accuracy in device-edge CNN inference systems.

Findings

AECNN compresses data by over 256x with only 4% accuracy loss

Outperforms state-of-the-art BottleNet++ in compression and accuracy

Enables faster inference under poor wireless conditions

Abstract

This paper studies the computational offloading of CNN inference in device-edge co-inference systems. Inspired by the emerging paradigm semantic communication, we propose a novel autoencoder-based CNN architecture (AECNN), for effective feature extraction at end-device. We design a feature compression module based on the channel attention method in CNN, to compress the intermediate data by selecting the most important features. To further reduce communication overhead, we can use entropy encoding to remove the statistical redundancy in the compressed data. At the receiver, we design a lightweight decoder to reconstruct the intermediate data through learning from the received compressed data to improve accuracy. To fasten the convergence, we use a step-by-step approach to train the neural networks obtained based on ResNet-50 architecture. Experimental results show that AECNN can compress the intermediate data by more than 256x with only about 4% accuracy loss, which outperforms the state-of-the-art work, BottleNet++. Compared to offloading inference task directly to edge server, AECNN can complete inference task earlier, in particular, under poor wireless channel condition, which highlights the effectiveness of AECNN in guaranteeing higher accuracy within time constraint.