An adaptive reduced basis ANOVA method forhigh-dimensional Bayesian inverse problems

doi:10.1016/j.jcp.2019.06.059

	An adaptive reduced basis ANOVA method forhigh-dimensional Bayesian inverse problems
	Liao, Qifeng1 ; Li, Jinglai 2
	2019-11
发表期刊	JOURNAL OF COMPUTATIONAL PHYSICS
ISSN	0021-9991
卷号	396 页码:364-380
发表状态	已发表
DOI	10.1016/j.jcp.2019.06.059
摘要	In Bayesian inverse problems sampling the posterior distribution is often a challenging task when the underlying models are computationally intensive. To this end, surrogates or reduced models are often used to accelerate the computation. However, in many practical problems, the parameter of interest can be of high dimensionality, which renders standard model reduction techniques infeasible. In this paper, we present an approach that employs the ANOVA decomposition method to reduce the model with respect to the unknown parameters, and the reduced basis method to reduce the model with respect to the physical parameters. Moreover, we provide an adaptive scheme within the MCMC iterations, to perform the ANOVA decomposition with respect to the posterior distribution. With numerical examples, we demonstrate that the proposed model reduction method can significantly reduce the computational cost of Bayesian inverse problems, without sacrificing much accuracy. (C) 2019 Elsevier Inc. All rights reserved.
关键词	ANOVA Reduced basis methods Bayesian inference Markov Chain Monte Carlo Inverse problems
URL	查看原文
收录类别	SCI ; SCIE ; EI
语种	英语
资助项目	[11771289]
WOS研究方向	Computer Science ; Physics
WOS类目	Computer Science, Interdisciplinary Applications ; Physics, Mathematical
WOS记录号	WOS:000481732600019
出版者	ACADEMIC PRESS INC ELSEVIER SCIENCE
EI入藏号	20201708548730
EI主题词	Analysis of variance (ANOVA) ; Bayesian networks ; Differential equations ; Dimensionality reduction ; Inference engines ; Markov chains ; Monte Carlo methods ; Numerical methods
EI分类号	Expert Systems:723.4.1 ; Mathematics:921 ; Statistical Methods:922 ; Mathematical Statistics:922.2
WOS关键词	PARTIAL-DIFFERENTIAL-EQUATIONS ; STOCHASTIC COLLOCATION ; MODEL-REDUCTION ; EXPANSIONS ; PARAMETER ; APPROXIMATION ; INFERENCE
原始文献类型	Article
引用统计
文献类型	期刊论文
条目标识符	https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/66390
专题	信息科学与技术学院_PI研究组_廖奇峰组
通讯作者	Li, Jinglai
作者单位	1.ShanghaiTech Univ, Sch Informat Sci & Technol, Shanghai 201210, Peoples R China 2.Univ Liverpool, Dept Math Sci, Liverpool L69 7XL, Merseyside, England
第一作者单位	信息科学与技术学院
第一作者的第一单位	信息科学与技术学院
推荐引用方式 GB/T 7714	Liao, Qifeng,Li, Jinglai. An adaptive reduced basis ANOVA method forhigh-dimensional Bayesian inverse problems[J]. JOURNAL OF COMPUTATIONAL PHYSICS,2019,396:364-380.
APA	Liao, Qifeng,&Li, Jinglai.(2019).An adaptive reduced basis ANOVA method forhigh-dimensional Bayesian inverse problems.JOURNAL OF COMPUTATIONAL PHYSICS,396,364-380.
MLA	Liao, Qifeng,et al."An adaptive reduced basis ANOVA method forhigh-dimensional Bayesian inverse problems".JOURNAL OF COMPUTATIONAL PHYSICS 396(2019):364-380.