Optimal design of large-scale nonlinear Bayesian inverse problems under model uncertainty

被引:3
作者
Alexanderian, Alen [1 ]
Nicholson, Ruanui [2 ]
Petra, Noemi [3 ]
机构
[1] North Carolina State Univ, Dept Math, Raleigh, NC 27695 USA
[2] Univ Auckland, Dept Engn Sci, Auckland, New Zealand
[3] Univ Calif Merced, Dept Appl Math, Merced, CA USA
基金
美国国家科学基金会;
关键词
optimal experimental design; sensor placement; Bayesian inverse problems; model uncertainty; Bayesian approximation error; A-OPTIMAL DESIGN; COMPUTATIONAL FRAMEWORK; EFFICIENT ALGORITHMS; SENSITIVITY-ANALYSIS; INFERENCE;
D O I
10.1088/1361-6420/ad602e
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
We consider optimal experimental design (OED) for Bayesian nonlinear inverse problems governed by partial differential equations (PDEs) under model uncertainty. Specifically, we consider inverse problems in which, in addition to the inversion parameters, the governing PDEs include secondary uncertain parameters. We focus on problems with infinite-dimensional inversion and secondary parameters and present a scalable computational framework for optimal design of such problems. The proposed approach enables Bayesian inversion and OED under uncertainty within a unified framework. We build on the Bayesian approximation error (BAE) approach, to incorporate modeling uncertainties in the Bayesian inverse problem, and methods for A-optimal design of infinite-dimensional Bayesian nonlinear inverse problems. Specifically, a Gaussian approximation to the posterior at the maximum a posteriori probability point is used to define an uncertainty aware OED objective that is tractable to evaluate and optimize. In particular, the OED objective can be computed at a cost, in the number of PDE solves, that does not grow with the dimension of the discretized inversion and secondary parameters. The OED problem is formulated as a binary bilevel PDE constrained optimization problem and a greedy algorithm, which provides a pragmatic approach, is used to find optimal designs. We demonstrate the effectiveness of the proposed approach for a model inverse problem governed by an elliptic PDE on a three-dimensional domain. Our computational results also highlight the pitfalls of ignoring modeling uncertainties in the OED and/or inference stages.
引用
收藏
页数:30
相关论文
共 44 条
[1]   Optimal Design of Large-scale Bayesian Linear Inverse Problems Under Reducible Model Uncertainty: Good to Know What You Don't Know [J].
Alexanderian, Alen ;
Petra, Noemi ;
Stadler, Georg ;
Sunseri, Isaac .
SIAM-ASA JOURNAL ON UNCERTAINTY QUANTIFICATION, 2021, 9 (01) :163-184
[2]   Optimal experimental design for infinite-dimensional Bayesian inverse problems governed by PDEs: a review [J].
Alexanderian, Alen .
INVERSE PROBLEMS, 2021, 37 (04)
[3]   On Bayesian A- and D-Optimal Experimental Designs in Infinite Dimensions [J].
Alexanderian, Alen ;
Gloor, Philip J. ;
Ghattas, Omar .
BAYESIAN ANALYSIS, 2016, 11 (03) :671-695
[4]   A FAST AND SCALABLE METHOD FOR A-OPTIMAL DESIGN OF EXPERIMENTS FOR INFINITE-DIMENSIONAL BAYESIAN NONLINEAR INVERSE PROBLEMS [J].
Alexanderian, Alen ;
Petra, Noemi ;
Stadler, Georg ;
Ghattas, Omar .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2016, 38 (01) :A243-A272
[5]   A-OPTIMAL DESIGN OF EXPERIMENTS FOR INFINITE-DIMENSIONAL BAYESIAN LINEAR INVERSE PROBLEMS WITH REGULARIZED l0-SPARSIFICATION [J].
Alexanderian, Alen ;
Petra, Noemi ;
Stadler, Georg ;
Ghattas, Omar .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2014, 36 (05) :A2122-A2148
[6]   Estimating nuisance parameters in inverse problems [J].
Aravkin, Aleksandr Y. ;
van Leeuwen, Tristan .
INVERSE PROBLEMS, 2012, 28 (11)
[7]  
ATKINSON AC, 1992, OPTIMUM EXPT DESIGNS
[8]   STOCHASTIC LEARNING APPROACH FOR BINARY OPTIMIZATION: APPLICATION TO BAYESIAN OPTIMAL DESIGN OF EXPERIMENTS* [J].
ATTIA, A. H. M. E. D. ;
LEYFFER, S. V. E. N. ;
MUNSON, T. O. D. D. S. .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2022, 44 (02) :B395-B427
[9]   OPTIMAL EXPERIMENTAL DESIGN FOR INVERSE PROBLEMS IN THE PRESENCE OF OBSERVATION CORRELATIONS [J].
Attia, Ahmed ;
Constantinescu, Emil .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2022, 44 (04) :A2808-A2842
[10]   Inferring the basal sliding coefficient field for the Stokes ice sheet model under rheological uncertainty [J].
Babaniyi, Olalekan ;
Nicholson, Ruanui ;
Villa, Umberto ;
Petra, Noemi .
CRYOSPHERE, 2021, 15 (04) :1731-1750