Fast Bayesian experimental design: Laplace-based importance sampling for the expected information gain
Name:
Fast_Bayesian_experimental_design_Laplace-based_im.pdf
Size:
5.952Mb
Format:
PDF
Description:
Preprint
Type
ArticleKAUST Department
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) DivisionApplied Mathematics and Computational Science Program
KAUST Grant Number
CRG3 Award Ref:2281CRG4 Award Ref:2584
Date
2018-02-19Preprint Posting Date
2017-10-10Online Publication Date
2018-02-19Print Publication Date
2018-06Permanent link to this record
http://hdl.handle.net/10754/626493
Metadata
Show full item recordAbstract
In calculating expected information gain in optimal Bayesian experimental design, the computation of the inner loop in the classical double-loop Monte Carlo requires a large number of samples and suffers from underflow if the number of samples is small. These drawbacks can be avoided by using an importance sampling approach. We present a computationally efficient method for optimal Bayesian experimental design that introduces importance sampling based on the Laplace method to the inner loop. We derive the optimal values for the method parameters in which the average computational cost is minimized for a specified error tolerance. We use three numerical examples to demonstrate the computational efficiency of our method compared with the classical double-loop Monte Carlo, and a single-loop Monte Carlo method that uses the Laplace approximation of the return value of the inner loop. The first demonstration example is a scalar problem that is linear in the uncertain parameter. The second example is a nonlinear scalar problem. The third example deals with the optimal sensor placement for an electrical impedance tomography experiment to recover the fiber orientation in laminate composites.Citation
Beck J, Dia BM, Espath LFR, Long Q, Tempone R (2018) Fast Bayesian experimental design: Laplace-based importance sampling for the expected information gain. Computer Methods in Applied Mechanics and Engineering 334: 523–553. Available: http://dx.doi.org/10.1016/j.cma.2018.01.053.Sponsors
The research reported in this publication was supported by funding from King Abdullah University of Science and Technology (KAUST); KAUST CRG3 Award Ref:2281 and the KAUST CRG4 Award Ref:2584.Publisher
Elsevier BVarXiv
1710.03500Additional Links
http://arxiv.org/abs/1710.03500v1http://arxiv.org/pdf/1710.03500v1
http://www.sciencedirect.com/science/article/pii/S0045782518300616
ae974a485f413a2113503eed53cd6c53
10.1016/j.cma.2018.01.053