Limitations of polynomial chaos expansions in the Bayesian solution of inverse problems

被引:35
作者
Lu, Fei [1 ,2 ]
Morzfeld, Matthias [1 ,2 ]
Tu, Xuemin [3 ]
Chorin, Alexandre J. [1 ,2 ]
机构
[1] Lawrence Berkeley Natl Lab, Berkeley, CA 94720 USA
[2] Univ Calif Berkeley, Dept Math, Berkeley, CA 94720 USA
[3] Univ Kansas, Dept Math, Lawrence, KS 66045 USA
基金
美国国家科学基金会;
关键词
Polynomial chaos expansion; Bayesian inverse problem; Monte Carlo sampling; UNCERTAINTY QUANTIFICATION; INFERENCE; EQUATIONS;
D O I
10.1016/j.jcp.2014.11.010
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
Polynomial chaos expansions are used to reduce the computational cost in the Bayesian solutions of inverse problems by creating a surrogate posterior that can be evaluated inexpensively. We show, by analysis and example, that when the data contain significant information beyond what is assumed in the prior, the surrogate posterior can be very different from the posterior, and the resulting estimates become inaccurate. One can improve the accuracy by adaptively increasing the order of the polynomial chaos, but the cost may increase too fast for this to be cost effective compared to Monte Carlo sampling without a surrogate posterior. (C) 2014 Elsevier Inc. All rights reserved.
引用
收藏
页码:138 / 147
页数:10
相关论文
共 32 条
  • [1] [Anonymous], 2010, SCI COMPUT
  • [2] [Anonymous], ARXIV13084640
  • [3] [Anonymous], CBMS NSF REG C SER A
  • [4] [Anonymous], STICHASTIC TOOLS MAT
  • [5] [Anonymous], 1997, THEORY FAST SOLVERS
  • [6] Implicit Particle Methods and Their Connection with Variational Data Assimilation
    Atkins, Ethan
    Morzfeld, Matthias
    Chorin, Alexandre J.
    [J]. MONTHLY WEATHER REVIEW, 2013, 141 (06) : 1786 - 1803
  • [7] Uncertainty reduction and characterization for complex environmental fate and transport models: An empirical Bayesian framework incorporating the stochastic response surface method
    Balakrishnan, S
    Roy, A
    Ierapetritou, MG
    Flach, GP
    Georgopoulos, PG
    [J]. WATER RESOURCES RESEARCH, 2003, 39 (12)
  • [8] Solution of inverse problems with limited forward solver evaluations: a Bayesian perspective
    Bilionis, I.
    Zabaras, N.
    [J]. INVERSE PROBLEMS, 2014, 30 (01)
  • [9] Branicki M, 2013, COMMUN MATH SCI, V11, P55
  • [10] GAUSSIAN FIELDS AND RANDOM FLOW
    CHORIN, AJ
    [J]. JOURNAL OF FLUID MECHANICS, 1974, 63 (MAR18) : 21 - 32