Mini-Tutorial 4: Leveraging Bayesian Methods to support Integrated Testing
Justin Krometis is a Research Assistant Professor in the Intelligent Systems Division of the Virginia Tech National Security Institute and an Affiliate Research Assistant Professor in the Virginia Tech Department of Mathematics. His research is in the development of theoretical and computational frameworks for Bayesian inference, particularly in high-dimensional regimes, and in the application of those methods to domain sciences ranging from fluids to geophysics to testing and evaluation. His areas of interest include statistical inverse problems, parameter estimation, machine learning, data science, and experimental design. Dr. Krometis holds a Ph.D. in mathematics, a M.S. in mathematics, a B.S. in mathematics, and a B.S. in physics, all from Virginia Tech.
Adam S. Ahmed is a Research Scientist at Metron, Inc. and is the technical lead for the Metron DOT&E effort. His research interests include applying novel Bayesian approaches to testing and evaluation, machine learning methods for small datasets as applied to undersea mine classification, and time series classification for continuous active sonar systems. Prior to Metron, he worked on the synthesis and measurement of skyrmion-hosting materials for next generation magnetic memory storage devices at The Ohio State University. Dr. Ahmed holds a Ph.D. and M.S. in physics from The Ohio State University, and a B.S. in physics from University of Illinois Urbana-Champaign.
This mini-tutorial will outline approaches to apply Bayesian methods to the test and evaluation process, from development of tests to interpretation of test results to translating that understanding into decision-making. We will begin by outlining the basic concepts that underlie the Bayesian approach to statistics and the potential benefits of applying that approach to test and evaluation. We will then walk through application to an example (notional) program, setting up data models and priors on the associated parameters, and interpreting the results. From there, techniques for integrating results from multiple stages of tests will be discussed, building understanding of system behavior as evidence accumulates. Finally, we will conclude by describing how Bayesian thinking can be used to translate information from test outcomes into requirements and decision-making. The mini-tutorial will assume some background in statistics but the audience need not have prior exposure to Bayesian methods.
Session Materials: https://dataworks.testscience.org/wp-...