Structural identifiability of generalized constraint neural network models for nonlinear regression

CASIA OpenIR > 09年以前成果

	Structural identifiability of generalized constraint neural network models for nonlinear regression
	Yang, Shuang-Hong 1,2; Hu, Bao-Gang1,2 ; Cournede, Paul-Henry 3
发表期刊	NEUROCOMPUTING
	2008-12-01
卷号	72 期号:1-3 页码:392-400
文章类型	Article
摘要	Identifiability becomes an essential requirement for learning machines when the models contain physically interpretable parameters. This paper presents two approaches to examining structural identifiability of the generalized constraint neural network (GCNN) models by viewing the model from two different perspectives. First, by taking the model as a static deterministic function, a functional framework is established, which can recognize deficient model and at the same time reparameterize it through a pairwise-mode symbolic examination. Second, by viewing the model as the mean function of an isotropic Gaussian conditional distribution, the algebraic approaches [E.A. Catchpole, B.J.T. Morgan, Detecting parameter redundancy, Biometrika 84 (1) (1997) 187-196] are extended to deal with multivariate nonlinear regression models through symbolically checking linear dependence of the derivative functional vectors. Examples are presented in which the proposed approaches are applied to GCNN nonlinear regression models that contain coupling physically interpretable parameters. (C) 2007 Elsevier B.V. All rights reserved.
关键词	Identifiability Parameter Redundancy Derivative Functional Vector Nonlinear Regression Hybrid Neural Network
WOS标题词	Science & Technology ; Technology
关键词[WOS]	PARAMETER REDUNDANCY ; SYMBOLIC COMPUTATION ; DISTINGUISHABILITY
收录类别	SCI
语种	英语
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000261643700045
引用统计	被引频次：11[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/9617
专题	09年以前成果
作者单位	1.Chinese Acad Sci, Inst Automat, NLPR, Beijing 100080, Peoples R China 2.Chinese Acad Sci, Inst Automat, LIAMA, Beijing 100080, Peoples R China 3.Ecole Cent Paris, Lab Appl Math & Syst, F-92295 Chatenay Malabry, France
第一作者单位	模式识别国家重点实验室; 中国科学院自动化研究所
推荐引用方式 GB/T 7714	Yang, Shuang-Hong,Hu, Bao-Gang,Cournede, Paul-Henry. Structural identifiability of generalized constraint neural network models for nonlinear regression[J]. NEUROCOMPUTING,2008,72(1-3):392-400.
APA	Yang, Shuang-Hong,Hu, Bao-Gang,&Cournede, Paul-Henry.(2008).Structural identifiability of generalized constraint neural network models for nonlinear regression.NEUROCOMPUTING,72(1-3),392-400.
MLA	Yang, Shuang-Hong,et al."Structural identifiability of generalized constraint neural network models for nonlinear regression".NEUROCOMPUTING 72.1-3(2008):392-400.