Bayesian Optimization as Black Box Method for Complex Engineering Applications
Abstract
The talk discusses black-box optimization methods, specifically random search approaches, for complex engineering applications. The speaker presents algorithms that focus on control, acceleration of the explore/exploit process, and scalability into high dimensions. The talk also explores the application of Bayesian optimization in the certification of safety-critical systems. The speaker introduces Part-X, a family of partitioning-informed Bayesian optimizers, and min-BO, an algorithm for identifying faults in systems with complex requirements. The talk concludes with future research directions to expand Bayesian optimization in embedding structure for complex engineered systems.
Bio
Giulia Pedrielli is an Assistant Professor for the School of Computing and Augmented Intelligence (SCAI) at Arizona State University. Her research focuses on the design and analysis of random algorithms for global optimization, with applications in manufacturing systems, bio-manufacturing, and life sciences. Her work is funded by organizations such as NSF, DHS, DARPA, Intel, and Lockheed Martin.
