Determining parameter identifiability from the optimization theory framework: A Kullback-Leibler divergence approach

	Determining parameter identifiability from the optimization theory framework: A Kullback-Leibler divergence approach
	Ran, Zhi-Yong 1; Hu, Bao-Gang
发表期刊	NEUROCOMPUTING
	2014-10-22
卷号	142 期号:2 页码:307-317
文章类型	Article
摘要	This paper reports an extension of the existing investigations on determining identifiability of statistical parameter models. By making use of the Kullback-Leibler divergence (KLD) in information theory, we cast the identifiability problem into the optimization theory framework. This is the first work that studies the identifiability problem from the optimization theory perspective which leads to connections in many areas of scientific research, e.g., identifiability theory, information theory and optimization theory. Within this new framework, we derive identifiability criteria according to the types of models. First, by formulating the identifiability problem of unconstrained parameter models as an unconstrained optimization problem, we derive identifiability criteria by checking the rank of the Hessian matrix of KLD. The resulting theorems extend the existing approaches and work in arbitrary statistical models. Second, by formulating the identifiability problem of parameter-constrained models as a constrained optimization problem, we derive a novel criterion which has a clear algebraic and geometric interpretation. Further, we discuss the pros/cons of the new framework from both theoretical and application viewpoints. Several model examples from the literature are presented to examine their identifiability property. (C) 2014 Elsevier B.V. All rights reserved.
关键词	Identifiability Optimization Theory Kullback-leibler Divergence Hessian Matrix Jacobian Matrix
WOS标题词	Science & Technology ; Technology
关键词[WOS]	NEURAL-NETWORK MODEL ; STRUCTURAL IDENTIFIABILITY ; INFORMATION CRITERION ; LEARNING MACHINES ; IDENTIFICATION ; CONSTRAINTS ; REGULARITY ; DYNAMICS
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:000340341400033
引用统计	被引频次：8[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/173211/2872
专题	多模态人工智能系统全国重点实验室_多媒体计算
作者单位	1.Chinese Acad Sci, NLPR, Beijing 100190, Peoples R China 2.Chinese Acad Sci, Inst Automat, LIAMA, Beijing 100190, Peoples R China
第一作者单位	模式识别国家重点实验室
推荐引用方式 GB/T 7714	Ran, Zhi-Yong,Hu, Bao-Gang. Determining parameter identifiability from the optimization theory framework: A Kullback-Leibler divergence approach[J]. NEUROCOMPUTING,2014,142(2):307-317.
APA	Ran, Zhi-Yong,&Hu, Bao-Gang.(2014).Determining parameter identifiability from the optimization theory framework: A Kullback-Leibler divergence approach.NEUROCOMPUTING,142(2),307-317.
MLA	Ran, Zhi-Yong,et al."Determining parameter identifiability from the optimization theory framework: A Kullback-Leibler divergence approach".NEUROCOMPUTING 142.2(2014):307-317.

条目包含的文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
Ran14b.pdf（563KB）	期刊论文	作者接受稿	开放获取	CC BY-NC-SA	浏览