Towards a Stable Quantized Convolutional Neural Networks: An Embedded Perspective

Motaz Al-Hami, Marcin Pietron, Raul Casas, Samer Hijazi, Piyush Kaul

2018

Abstract

Nowadays, convolutional neural network (CNN) plays a major role in the embedded computing environment. Ability to enhance the CNN implementation and performance for embedded devices is an urgent demand. Compressing the network layers parameters and outputs into a suitable precision formats would reduce the required storage and computation cycles in embedded devices. Such enhancement can drastically reduce the consumed power and the required resources, and ultimately reduces cost. In this article, we propose several quantization techniques for quantizing several CNN networks. With a minor degradation of the floating-point performance, the presented quantization methods are able to produce a stable performance fixed-point networks. A precise fixed point calculation for coefficients, input/output signals and accumulators are considered in the quantization process.

Paper Citation

in Harvard Style

Al-Hami M., Pietron M., Casas R., Hijazi S. and Kaul P. (2018). Towards a Stable Quantized Convolutional Neural Networks: An Embedded Perspective.In Proceedings of the 10th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART, ISBN 978-989-758-275-2, pages 573-580. DOI: 10.5220/0006651305730580

in Bibtex Style

@conference{icaart18,
author={Motaz Al-Hami and Marcin Pietron and Raul Casas and Samer Hijazi and Piyush Kaul},
title={Towards a Stable Quantized Convolutional Neural Networks: An Embedded Perspective},
booktitle={Proceedings of the 10th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,},
year={2018},
pages={573-580},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006651305730580},
isbn={978-989-758-275-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 10th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,
TI - Towards a Stable Quantized Convolutional Neural Networks: An Embedded Perspective
SN - 978-989-758-275-2
AU - Al-Hami M.
AU - Pietron M.
AU - Casas R.
AU - Hijazi S.
AU - Kaul P.
PY - 2018
SP - 573
EP - 580
DO - 10.5220/0006651305730580