CondenseNet: An Efficient DenseNet using Learned Group Convolutions

被引：541

作者：

Huang, Gao ^{[1
]}

Liu, Shichen ^{[2
]}

van der Maaten, Laurens ^{[3
]}

Weinberger, Kilian Q. ^{[1
]}

机构：

[1] Cornell Univ, Ithaca, NY 14853 USA

[2] Tsinghua Univ, Beijing, Peoples R China

[3] Facebook AI Res, Menlo Pk, CA USA

来源：

2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR) | 2018年

基金：

美国国家科学基金会;

关键词：

D O I：

10.1109/CVPR.2018.00291

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Deep neural networks are increasingly used on mobile devices, where computational resources are limited. In this paper we develop CondenseNet, a novel network architecture with unprecedented efficiency. It combines dense connectivity with a novel module called learned group convolution. The dense connectivity facilitates feature re-use in the network, whereas learned group convolutions remove connections between layers for which this feature re-use is superfluous. At test time, our model can be implemented using standard group convolutions, allowing for efficient computation in practice. Our experiments show that Condense-Nets are far more efficient than state-of-the-art compact convolutional networks such as ShuffleNets.

引用

页码：2752 / 2761

页数：10