Simulation-based optimal Bayesian experimental design for nonlinear systems

Handle URI:
http://hdl.handle.net/10754/599624
Title:
Simulation-based optimal Bayesian experimental design for nonlinear systems
Authors:
Huan, Xun; Marzouk, Youssef M.
Abstract:
The optimal selection of experimental conditions is essential to maximizing the value of data for inference and prediction, particularly in situations where experiments are time-consuming and expensive to conduct. We propose a general mathematical framework and an algorithmic approach for optimal experimental design with nonlinear simulation-based models; in particular, we focus on finding sets of experiments that provide the most information about targeted sets of parameters.Our framework employs a Bayesian statistical setting, which provides a foundation for inference from noisy, indirect, and incomplete data, and a natural mechanism for incorporating heterogeneous sources of information. An objective function is constructed from information theoretic measures, reflecting expected information gain from proposed combinations of experiments. Polynomial chaos approximations and a two-stage Monte Carlo sampling method are used to evaluate the expected information gain. Stochastic approximation algorithms are then used to make optimization feasible in computationally intensive and high-dimensional settings. These algorithms are demonstrated on model problems and on nonlinear parameter inference problems arising in detailed combustion kinetics. © 2012 Elsevier Inc.
Citation:
Huan X, Marzouk YM (2013) Simulation-based optimal Bayesian experimental design for nonlinear systems. Journal of Computational Physics 232: 288–317. Available: http://dx.doi.org/10.1016/j.jcp.2012.08.013.
Publisher:
Elsevier BV
Journal:
Journal of Computational Physics
Issue Date:
Jan-2013
DOI:
10.1016/j.jcp.2012.08.013
Type:
Article
ISSN:
0021-9991
Sponsors:
The authors would like to acknowledge support from the KAUST Global Research Partnership and from the US Department of Energy, Office of Science, Office of Advanced Scientific Computing Research (ASCR) under Grant No. DE-SC0003908.
Appears in Collections:
Publications Acknowledging KAUST Support

Full metadata record

DC FieldValue Language
dc.contributor.authorHuan, Xunen
dc.contributor.authorMarzouk, Youssef M.en
dc.date.accessioned2016-02-28T06:06:05Zen
dc.date.available2016-02-28T06:06:05Zen
dc.date.issued2013-01en
dc.identifier.citationHuan X, Marzouk YM (2013) Simulation-based optimal Bayesian experimental design for nonlinear systems. Journal of Computational Physics 232: 288–317. Available: http://dx.doi.org/10.1016/j.jcp.2012.08.013.en
dc.identifier.issn0021-9991en
dc.identifier.doi10.1016/j.jcp.2012.08.013en
dc.identifier.urihttp://hdl.handle.net/10754/599624en
dc.description.abstractThe optimal selection of experimental conditions is essential to maximizing the value of data for inference and prediction, particularly in situations where experiments are time-consuming and expensive to conduct. We propose a general mathematical framework and an algorithmic approach for optimal experimental design with nonlinear simulation-based models; in particular, we focus on finding sets of experiments that provide the most information about targeted sets of parameters.Our framework employs a Bayesian statistical setting, which provides a foundation for inference from noisy, indirect, and incomplete data, and a natural mechanism for incorporating heterogeneous sources of information. An objective function is constructed from information theoretic measures, reflecting expected information gain from proposed combinations of experiments. Polynomial chaos approximations and a two-stage Monte Carlo sampling method are used to evaluate the expected information gain. Stochastic approximation algorithms are then used to make optimization feasible in computationally intensive and high-dimensional settings. These algorithms are demonstrated on model problems and on nonlinear parameter inference problems arising in detailed combustion kinetics. © 2012 Elsevier Inc.en
dc.description.sponsorshipThe authors would like to acknowledge support from the KAUST Global Research Partnership and from the US Department of Energy, Office of Science, Office of Advanced Scientific Computing Research (ASCR) under Grant No. DE-SC0003908.en
dc.publisherElsevier BVen
dc.subjectBayesian inferenceen
dc.subjectChemical kineticsen
dc.subjectNonlinear experimental designen
dc.subjectOptimal experimental designen
dc.subjectShannon informationen
dc.subjectStochastic approximationen
dc.subjectUncertainty quantificationen
dc.titleSimulation-based optimal Bayesian experimental design for nonlinear systemsen
dc.typeArticleen
dc.identifier.journalJournal of Computational Physicsen
dc.contributor.institutionMassachusetts Institute of Technology, Cambridge, United Statesen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.