Complex hybrid weighted pruning method for accelerating convolutional neural networks

Abstract The increasing interest in filter pruning of convolutional neural networks stems from its inherent ability to effectively compress and accelerate these networks. Currently, filter pruning is mainly divided into two schools: norm-based and relation-based. These methods aim to selectively rem...

Full description

Bibliographic Details
Published in:Scientific Reports
Main Authors: Xu Geng, Jinxiong Gao, Yonghui Zhang, Dingtan Xu
Format: Article
Language:English
Published: Nature Portfolio 2024-03-01
Online Access:https://doi.org/10.1038/s41598-024-55942-5