Content area

Abstract

Channel pruning is a method to compress convolutional neural networks, which can significantly reduce the number of model parameters and the computational amount. Current methods that focus on the internal parameters of a model and feature mapping information rely on artificially set a priori criteria or reflect filter attributes by partial feature mapping, which lack the ability to analyze and discriminate the channel feature extraction and ignore the basic reasons for the similarity of the channels. This study developed a pruning method based on similar structural features of channels, called SSF. This method focuses on analysing the ability to extract similar features between channels and exploring the characteristics of channels producing similar feature mapping. First, adaptive threshold coding was introduced to numerically transform the channel characteristics into structural features, and channels with similar coding results could generate highly similar feature mapping. Secondly, the spatial distance was calculated for the structural features matrix to obtain the similarity between channels. Moreover, in order to keep rich channel classes in the pruned network, different class cuts were made on the basis of similarity to randomly remove some of the channels. Thirdly, considering the differences in the overall similarity of different layers, this study determined the appropriate pruning ratio for different layers on the basis of the channel dispersion degree reflected by the similarity. Finally, extensive experiments were conducted on image classification tasks, and the experimental results demonstrated the superiority of the SSF method over many existing techniques. On ILSVRC-2012, the SSF method reduced the floating-point operations (FLOPs) of the ResNet-50 model by 57.70% while reducing the Top-1 accuracy only by 1.01%.

Details

Title
Channel pruning method driven by similarity of feature extraction capability
Publication title
Soft Computing; Heidelberg
Volume
29
Issue
2
Pages
1207-1226
Publication year
2025
Publication date
Jan 2025
Publisher
Springer Nature B.V.
Place of publication
Heidelberg
Country of publication
Netherlands
ISSN
14327643
e-ISSN
14337479
Source type
Scholarly Journal
Language of publication
English
Document type
Journal Article
Publication history
 
 
Online publication date
2025-02-08
Milestone dates
2025-01-21 (Registration); 2024-11-08 (Accepted)
Publication history
 
 
   First posting date
08 Feb 2025
ProQuest document ID
3168165096
Document URL
https://www.proquest.com/scholarly-journals/channel-pruning-method-driven-similarity-feature/docview/3168165096/se-2?accountid=208611
Copyright
Copyright Springer Nature B.V. Jan 2025
Last updated
2025-02-19
Database
ProQuest One Academic