Recent studies have shown the latency and energy consumption of deep neural networks can be significantly improved by splitting the network between the mobile device and cloud. This paper introduces a new deep learning architecture, called BottleNet, for reducing the feature size needed to be sent to the cloud. Furthermore, we propose a training method for compensating for the potential accuracy loss due to the lossy compression of features before transmitting them to the cloud. BottleNet achieves on average 30x improvement in end-to-end latency and 40x improvement in mobile energy consumption compared to the cloud-only approach with negligible accuracy loss.
@article{arxiv.1902.01000,
title = {BottleNet: A Deep Learning Architecture for Intelligent Mobile Cloud Computing Services},
author = {Amir Erfan Eshratifar and Amirhossein Esmaili and Massoud Pedram},
journal= {arXiv preprint arXiv:1902.01000},
year = {2019}
}
Comments
arXiv admin note: text overlap with arXiv:1902.00147