OICSR: Out-In-Channel Sparsity Regularization for Compact Deep Neural Networks

Channel pruning can significantly accelerate and compress deep neural networks. Many channel pruning works utilize structured sparsity regularization to zero out all the weights in some channels and automatically obtain structure-sparse network in training stage. However, these methods apply structu...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2019-07
Hauptverfasser:	Li, Jiashi, Qi, Qi, Wang, Jingyu, Ge, Ce, Li, Yujian, Zhangzhang Yue, Sun, Haifeng
Format:	Artikel
Sprache:	eng
Schlagworte:	Channels Greedy algorithms Iterative methods Model accuracy Neural networks Pruning Regularization Sparsity Training
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!