A nonlinear convolution neural network quantization method

Lou Zheng; Fan Lei; Tao Weisong; Xu Chao

doi:10.1117/12.3051024

27 September 2024 A nonlinear convolution neural network quantization method

Lou Zheng, Fan Lei, Tao Weisong, Xu Chao

Proceedings Volume 13281, International Conference on Cloud Computing, Performance Computing, and Deep Learning (CCPCDL 2024); 132810W (2024) https://doi.org/10.1117/12.3051024
Event: International Conference on Cloud Computing, Performance Computing, and Deep Learning, 2024, Zhengzhou, China

Abstract

In today's world, convolution neural networks possess tremendous potential and exhibit remarkable recognition rates in computer vision. However, neural network models are often so large that they are not well-suited for mobile devices, making model compression particularly important. Especially in scenarios with limited hardware resources such as the National Grid, model quantification is essential. This paper introduces an INT8 quantization method for convolution neural networks, which transforms the weights and input/output of the convolution neural network into the logarithmic domain through a non-linear formula. This method provides a better representation of the distribution of small values, resulting in less accuracy loss. Meanwhile, model calibration requires only a limited amount of image data, which reduces the time taken for model calibration while ensuring minimal accuracy loss. Experimental results show that our quantization method results in an accuracy loss of less than 1% on ImageNet and CIFAR.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Lou Zheng, Fan Lei, Tao Weisong, and Xu Chao "A nonlinear convolution neural network quantization method", Proc. SPIE 13281, International Conference on Cloud Computing, Performance Computing, and Deep Learning (CCPCDL 2024), 132810W (27 September 2024); https://doi.org/10.1117/12.3051024

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
8 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Quantization

Convolution

Neural networks

Data modeling

Education and training

Calibration

Performance modeling

Show All Keywords

Keywords/Phrases

Search In:

Publication Years