Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

TOWARDS THINNER CONVOLUTIONAL NEURAL NETWORKS THROUGH GRADUALLY GLOBAL PRUNING

Abstract: 

Deep network pruning is an effective method to reduce the storage and computation cost of deep neural networks when applying them to resource-limited devices. Among many pruning granularity, neuron level pruning will remove redundant neurons and filters in the model and result in thinner networks. In this paper, we propose a gradually global pruning scheme for neuron level pruning. In each pruning step, a small percent of neurons were selected and dropped across all layers in the model. We also propose a simple method to eliminate the biases in evaluating the importance of neurons to make the scheme feasible. Compared with layer-wise pruning scheme, our scheme avoid the difficulty in determining the redundancy in each layer and is more effective for deep networks. Our scheme would automatically find a thinner sub-network in original network under a given performance.

up
0 users have voted:

Paper Details

Authors:
Ce Zhu, Zhiqiang Xia, Qi Guo, Yipeng Liu
Submitted On:
13 September 2017 - 12:37pm
Short Link:
Type:
Poster
Event:
Presenter's Name:
Zhengtao Wang
Paper Code:
ICIP1701
Document Year:
2017
Cite

Document Files

ICIP1701

(10 downloads)

Subscribe

[1] Ce Zhu, Zhiqiang Xia, Qi Guo, Yipeng Liu, "TOWARDS THINNER CONVOLUTIONAL NEURAL NETWORKS THROUGH GRADUALLY GLOBAL PRUNING", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1979. Accessed: Sep. 24, 2017.
@article{1979-17,
url = {http://sigport.org/1979},
author = {Ce Zhu; Zhiqiang Xia; Qi Guo; Yipeng Liu },
publisher = {IEEE SigPort},
title = {TOWARDS THINNER CONVOLUTIONAL NEURAL NETWORKS THROUGH GRADUALLY GLOBAL PRUNING},
year = {2017} }
TY - EJOUR
T1 - TOWARDS THINNER CONVOLUTIONAL NEURAL NETWORKS THROUGH GRADUALLY GLOBAL PRUNING
AU - Ce Zhu; Zhiqiang Xia; Qi Guo; Yipeng Liu
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1979
ER -
Ce Zhu, Zhiqiang Xia, Qi Guo, Yipeng Liu. (2017). TOWARDS THINNER CONVOLUTIONAL NEURAL NETWORKS THROUGH GRADUALLY GLOBAL PRUNING. IEEE SigPort. http://sigport.org/1979
Ce Zhu, Zhiqiang Xia, Qi Guo, Yipeng Liu, 2017. TOWARDS THINNER CONVOLUTIONAL NEURAL NETWORKS THROUGH GRADUALLY GLOBAL PRUNING. Available at: http://sigport.org/1979.
Ce Zhu, Zhiqiang Xia, Qi Guo, Yipeng Liu. (2017). "TOWARDS THINNER CONVOLUTIONAL NEURAL NETWORKS THROUGH GRADUALLY GLOBAL PRUNING." Web.
1. Ce Zhu, Zhiqiang Xia, Qi Guo, Yipeng Liu. TOWARDS THINNER CONVOLUTIONAL NEURAL NETWORKS THROUGH GRADUALLY GLOBAL PRUNING [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1979