Notice

This is not the latest version of this item. The latest version can be found at: https://repository.kaust.edu.sa/handle/10754/626493

Show simple item record

dc.contributor.authorBeck, Joakimen
dc.contributor.authorDia, Ben Mansouren
dc.contributor.authorEspath, Luis FRen
dc.contributor.authorLong, Quanen
dc.contributor.authorTempone, Raulen
dc.date.accessioned2017-12-28T07:32:13Z
dc.date.available2017-12-28T07:32:13Z
dc.date.issued2017-10-10en
dc.identifier.urihttp://hdl.handle.net/10754/626493.1
dc.description.abstractIn 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 according to the desired 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 more recent single-loop Monte Carlo method that uses the Laplace method as an approximation of the return value of the inner loop. The first 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.en
dc.publisherarXiven
dc.relation.urlhttp://arxiv.org/abs/1710.03500v1en
dc.relation.urlhttp://arxiv.org/pdf/1710.03500v1en
dc.rightsArchived with thanks to arXiven
dc.titleFast Bayesian experimental design: Laplace-based importance sampling for the expected information gainen
dc.typePreprinten
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.eprint.versionPre-printen
dc.contributor.institutionCIPR, College of Petroleum Engineering and Geosciences, King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabiaen
dc.contributor.institutionUnited Technologies Research Center, East Hartford, CT, 06108, United Statesen
dc.identifier.arxividarXiv:1710.03500en
kaust.personBeck, Joakim
kaust.personEspath, Luis FR
kaust.personTempone, Raul
kaust.grant.number2281en
kaust.grant.number2584en


Files in this item

Thumbnail
Name:
1710.03500v1.pdf
Size:
6.316Mb
Format:
PDF
Description:
Preprint

This item appears in the following Collection(s)

Show simple item record

VersionItemEditorDateSummary

*Selected version