Abstract:
Realistic state-based discrete-event simulation models are often quite complex. The complexity frequently manifests in models that (a) contain many input variables whose values are difficult to determine precisely, and (b) take a relatively long time to solve. Traditionally, models that have many input variables whose values are not well-known are understood using sensitivity analysis (SA) and uncertainty quantification (UQ). However, it can be prohibitively time consuming to perform SA and UQ. In this work, we present a novel ML-based approach we developed for performing faster SA and UQ on a metamodel composed of a stacked ensemble of regressors that emulates the behavior of the base model. We demonstrate the approach using previously published models as test cases, showing that the metamodels are several orders of magnitude faster than the corresponding base models, more accurate than existing approaches, and amenable to SA and UQ.
Biosketch:
Michael Rausch is a postdoc at the University of Illinois at Urbana-Champaign. His research interests are at the intersection of cybersecurity modeling, machine learning, and simulation. He has published multiple papers in this area, including winning a Best Paper Award at QEST2020. Dr. Rausch has participated in research projects as an intern at Sandia National Laboratories, the University of California-Berkeley, and Colorado State University. He earned his PhD from the University of Illinois at Urbana-Champaign, where he was advised by William H. Sanders.
