Title: Application of Bayesian parameter inference to model the permeation of Gallium through Aluminum grain boundaries.
Abstract: We are studying the application of Bayesian parameter inference in modeling grain boundary physics. The permeation of liquid Gallium through Aluminum grain boundaries is being used as a model problem for this study. A modified Cahn-Hilliard equation is used to model the permeation of Gallium. Aluminum grain boundaries are generated using molecular dynamics simulations. Parameters of the Cahn-Hilliard model will be estimated using data found in existing literature in combination with Bayesian parameter inference.
Title: A Layered Multiple Importance Sampling Scheme for Focused Optimal Bayesian Experimental Design
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 costly. We propose an information theoretic framework for focused experimental design with simulation-based models, with the goal of maximizing information gain in targeted subsets of model parameters. A novel layered multiple importance sampling technique is used to efficiently evaluate the expected information gain to make optimization feasible for computationally intensive and high-dimensional problems.