Multi-feature fusion deep networks

doi:10.1016/j.neucom.2016.08.059

	Multi-feature fusion deep networks
	Ma, Gang 1,2; Yang, Xi 1; Zhang, Bo 1,3; Shi, Zhongzhi 1
	2016-12-19
发表期刊	NEUROCOMPUTING
ISSN	0925-2312
卷号	218 页码:164-171
摘要	In this paper, we propose a novel deep networks, multi-feature fusion deep networks (MFFDN), based on denoising autoencoder. MFFDN significantly reduces the classification error while giving the interpretability of the hidden-layer feature representation in learning process. Comparing with the traditional denoising autoencoder, MFFDN mainly shows the following advantages: (1) minimally retaining a certain amount of "information" constrained to a given form about its input; (2) explicitly interpreting the meaning of the feature representation in one hidden layer; (3) enhancing discriminativeness of the whole networks. At last, the experiments analysis on MNIST, CIFAR-10 and SVHN prove the state-of-the-art performance improvement of MFFDN with the advantages minimally retaining "information" constraint and the interpreted hidden feature representation. (C) 2016 Elsevier B.V. All rights reserved.
关键词	Deep networks Denoising autoencoder Interpretability Discriminativeness
DOI	10.1016/j.neucom.2016.08.059
收录类别	SCI
语种	英语
资助项目	National Basic Research Program of China (973)[2013CB329502] ; National Natural Science Foundation of China[61035003] ; National Natural Science Foundation of China[61202212] ; National Science and Technology Support Program[2012BA107B02] ; Natural Science Foundation of Jiangsu Province[BK20160276]
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000388053700018
出版者	ELSEVIER SCIENCE BV
引用统计	被引频次：32[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/7923
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Ma, Gang
作者单位	1.Chinese Acad Sci, Key Lab Intelligent Informat Proc, Inst Comp Technol, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing 100190, Peoples R China 3.China Univ Min & Technol, Sch Comp Sci & Technol, Xuzhou 221116, Peoples R China
推荐引用方式 GB/T 7714	Ma, Gang,Yang, Xi,Zhang, Bo,et al. Multi-feature fusion deep networks[J]. NEUROCOMPUTING,2016,218:164-171.
APA	Ma, Gang,Yang, Xi,Zhang, Bo,&Shi, Zhongzhi.(2016).Multi-feature fusion deep networks.NEUROCOMPUTING,218,164-171.
MLA	Ma, Gang,et al."Multi-feature fusion deep networks".NEUROCOMPUTING 218(2016):164-171.