Suscripción a Biblioteca: Guest
Portal Digitalde Biblioteca Digital eLibros Revistas Referencias y Libros de Ponencias Colecciones
International Journal for Uncertainty Quantification
Factor de Impacto: 4.911 Factor de Impacto de 5 años: 3.179 SJR: 1.008 SNIP: 0.983 CiteScore™: 5.2

ISSN Imprimir: 2152-5080
ISSN En Línea: 2152-5099

Acceso abierto

International Journal for Uncertainty Quantification

DOI: 10.1615/Int.J.UncertaintyQuantification.2014006730
pages 479-510

GRADIENT-BASED STOCHASTIC OPTIMIZATION METHODS IN BAYESIAN EXPERIMENTAL DESIGN

Xun Huan
Sandia National Laboratories, 7011 East Ave, MS 9051, Livermore, CA 94550, USA; Department of Mechanical Engineering, University of Michigan, Ann Arbor, MI 48109, USA
Youssef Marzouk
Department of Aeronautics and Astronautics, Massachusetts Institute of Technology, 77 Massachusetts Ave, Room 33-305 Cambridge, MA 02139 USA

SINOPSIS

Optimal experimental design (OED) seeks experiments expected to yield the most useful data for some purpose. In practical circumstances where experiments are time-consuming or resource-intensive, OED can yield enormous savings. We pursue OED for nonlinear systems from a Bayesian perspective, with the goal of choosing experiments that are optimal for parameter inference. Our objective in this context is the expected information gain in model parameters, which in general can only be estimated using Monte Carlo methods. Maximizing this objective thus becomes a stochastic optimization problem. This paper develops gradient-based stochastic optimization methods for the design of experiments on a continuous parameter space. Given a Monte Carlo estimator of expected information gain, we use infinitesimal perturbation analysis to derive gradients of this estimator.We are then able to formulate two gradient-based stochastic optimization approaches: (i) Robbins-Monro stochastic approximation, and (ii) sample average approximation combined with a deterministic quasi-Newton method. A polynomial chaos approximation of the forward model accelerates objective and gradient evaluations in both cases.We discuss the implementation of these optimization methods, then conduct an empirical comparison of their performance. To demonstrate design in a nonlinear setting with partial differential equation forward models, we use the problem of sensor placement for source inversion. Numerical results yield useful guidelines on the choice of algorithm and sample sizes, assess the impact of estimator bias, and quantify tradeoffs of computational cost versus solution quality and robustness.


Articles with similar content:

YIELD OPTIMIZATION BASED ON ADAPTIVE NEWTON-MONTE CARLO AND POLYNOMIAL SURROGATES
International Journal for Uncertainty Quantification, Vol.10, 2020, issue 4
Niklas Georg, Mona Fuhrländer, Sebastian Schöps, Ulrich Römer
A MULTI-FIDELITY NEURAL NETWORK SURROGATE SAMPLING METHOD FOR UNCERTAINTY QUANTIFICATION
International Journal for Uncertainty Quantification, Vol.10, 2020, issue 4
Mohammad Motamed
QUANTIFICATION OF UNCERTAINTY FROM HIGH-DIMENSIONAL SCATTERED DATA VIA POLYNOMIAL APPROXIMATION
International Journal for Uncertainty Quantification, Vol.4, 2014, issue 3
Lionel Mathelin
ERROR AND UNCERTAINTY QUANTIFICATION AND SENSITIVITY ANALYSIS IN MECHANICS COMPUTATIONAL MODELS
International Journal for Uncertainty Quantification, Vol.1, 2011, issue 2
Sankaran Mahadevan, Bin Liang
BAYESIAN MULTISCALE FINITE ELEMENT METHODS. MODELING MISSING SUBGRID INFORMATION PROBABILISTICALLY
International Journal for Multiscale Computational Engineering, Vol.15, 2017, issue 2
Wing Tat Leung, B. Mallick, Yalchin Efendiev, N. Guha, V. H. Hoang, S. W. Cheung