A New Approach to Stochastic Inverse Problems for Scientific Inference
Tuesday, February 23, 2016 - 9:00am - 9:45am
The stochastic inverse problem for a physics model for determination of parameter values from observational data on the output of the model forms the core of scientific inference and engineering design. We describe a recently developed formulation and solution methodology that is based on rigorous measure theory and a generalization of a contour map. Advantages of this approach include avoiding the introduction of ad hoc statistics models, unverifiable assumptions, and alterations of the model like regularization. We present a high-dimensional application to determination of parameter fields in storm surge models. We conclude with recent work on defining a notion of condition for stochastic inverse problems and the use in choosing observable quantities.