An efficient channel pruning algorithm for automatic compression and acceleration of neural network models

Wei Xie; Xiaobo Feng

doi:10.1117/12.2660854

2 February 2023 An efficient channel pruning algorithm for automatic compression and acceleration of neural network models

Wei Xie, Xiaobo Feng

Author Affiliations +

Proceedings Volume 12462, Third International Symposium on Computer Engineering and Intelligent Communications (ISCEIC 2022); 124622S (2023) https://doi.org/10.1117/12.2660854
Event: International Symposium on Computer Engineering and Intelligent Communications (ISCEIC 2022), 2022, Xi'an, China

Abstract

The Convolutional Neural Network (CNN) enables deep neural networks to be deployed to resource-constrained mobile devices via model compression and acceleration. At present, channel pruning methods select channels based on channel importance or designed regularization, which are suboptimal pruning and cannot be automated. In this paper, a channel pruning algorithm is proposed to get the optimal pruned structure via automatic searching. By setting the super-parameter constraint set, the combination number of pruning structures is reduced. The number of channels for each layer of the CNN is determined using the sparrow search algorithm, and the optimal pruned structure of the model is found. The results of extensive experiments show that the proposed method can improve the model's parameter compression ratio and reduce the number of FLOPS within the acceptable range of model accuracy loss.

Citation Download Citation

Wei Xie and Xiaobo Feng "An efficient channel pruning algorithm for automatic compression and acceleration of neural network models", Proc. SPIE 12462, Third International Symposium on Computer Engineering and Intelligent Communications (ISCEIC 2022), 124622S (2 February 2023); https://doi.org/10.1117/12.2660854

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