pith. sign in

arxiv: 1108.4146 · v3 · pith:SHYGGT7Anew · submitted 2011-08-20 · 📊 stat.ML · stat.CO· stat.ME

Simulation-based optimal Bayesian experimental design for nonlinear systems

classification 📊 stat.ML stat.COstat.ME
keywords informationexperimentalexperimentsnonlinearoptimalalgorithmsbayesiandata
0
0 comments X p. Extension
pith:SHYGGT7A Add to your LaTeX paper What is a Pith Number?
\usepackage{pith}
\pithnumber{SHYGGT7A}

Prints a linked pith:SHYGGT7A badge after your title and writes the identifier into PDF metadata. Compiles on arXiv with no extra files. Learn more

read the original 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 estimation problems arising in detailed combustion kinetics.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.