图书馆订阅: Guest
Begell Digital Portal Begell 数字图书馆 电子图书 期刊 参考文献及会议录 研究收集
国际不确定性的量化期刊
影响因子: 3.259 5年影响因子: 2.547 SJR: 0.531 SNIP: 0.8 CiteScore™: 1.52

ISSN 打印: 2152-5080
ISSN 在线: 2152-5099

Open Access

国际不确定性的量化期刊

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

GRADIENT-BASED STOCHASTIC OPTIMIZATION METHODS IN BAYESIAN EXPERIMENTAL DESIGN

Xun Huan
University of Michigan
Youssef Marzouk
Department of Aeronautics and Astronautics, Massachusetts Institute of Technology, 77 Massachusetts Ave, Room 33-305 Cambridge, MA 02139 USA

ABSTRACT

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:

DATA-FREE INFERENCE OF UNCERTAIN PARAMETERS IN CHEMICAL MODELS
International Journal for Uncertainty Quantification, Vol.4, 2014, issue 2
Bert J. Debusschere, Robert D. Berry, Habib N. Najm, Cosmin Safta, Khachik Sargsyan
EFFECTIVE SAMPLING SCHEMES FOR BEHAVIOR DISCRIMINATION IN NONLINEAR SYSTEMS
International Journal for Uncertainty Quantification, Vol.4, 2014, issue 6
Gregery T. Buzzard, Vu Dinh, Ann E. Rundell
ORTHOGONAL BASES FOR POLYNOMIAL REGRESSION WITH DERIVATIVE INFORMATION IN UNCERTAINTY QUANTIFICATION
International Journal for Uncertainty Quantification, Vol.1, 2011, issue 4
Oleg Roderick, Mihai Anitescu, Fred Hickernell, Yiou Li
Closed-Form and Iterative Solutions to the Global Positioning System Problem
Journal of Automation and Information Sciences, Vol.34, 2002, issue 2
Marco Gatti, Salvatore Ponte, Nikolla Crocetto
INVERSE DETERMINATION OF SPATIALLY VARYING HEAT CAPACITY AND THERMAL CONDUCTIVITY IN ARBITRARY 2D OBJECTS
ICHMT DIGITAL LIBRARY ONLINE, Vol.0, 2017, issue
George S. Dulikravich, Sohail R. Reddy, S. M. Javad Zeidi