Speaker: Bruno Sudret, ETH Zurich
Date & Time: Friday, 23 January, 12:15–13:45
Location: MZ 2
Title: Sparse polynomial chaos expansions for uncertainty quantification and sensitivity analysis
Abstract:
Computational models play a central role in modern engineering and applied sciences, enabling the analysis, prediction, and optimization of complex systems. However, when uncertainties in model inputs, operating conditions, or environmental factors must be taken into account, the computational cost of traditional approaches such as Monte Carlo simulation for uncertainty quantification quickly becomes prohibitive. Surrogate models provide an efficient alternative by constructing accurate approximations of expensive simulators using a limited number of strategically selected model evaluations. These models can drastically reduce computational effort while retaining sufficient accuracy for uncertainty quantification tasks. This talk introduces the fundamental concepts of surrogate modelling and discusses their connections to machine learning methods. Particular emphasis is placed on polynomial chaos expansions, with a focus on sparse formulations designed to address high-dimensional problems. Recent methodological developments for dynamical systems and practical applications to global sensitivity analysis will also be presented.
Biography:
Bruno Sudret is a professor of Risk, Safety and Uncertainty quantification at ETH Zurich since 2012. His teaching and research interests are computational methods for uncertainty quantification, reliability and sensitivity analysis, Bayesian approaches for model calibration and reliability-based design optimization, among others.
B. Sudret received a master’s of science from the Ecole Polytechnique (France) in 1993. He then obtained a master’s degree and a Ph.D in civil engineering from the Ecole Nationale des Ponts et Chaussées (France) in 1996 and 1999, respectively. Dr. Sudret has been working in probabilistic engineering mechanics and uncertainty quantification for engineering systems since 2000: first as a post-doctoral fellow at the University of Berkeley (California), then as a researcher at EDF R&D (the French world leader in nuclear power generation) where he was the head of a group specialized in probabilistic engineering mechanics (2001-2008). From 2008 to 2011 he has worked as the Director of Research and Strategy at Phimeca Engineering (France).
B. Sudret is the author and co-author of more than 350 publications in journal and conference proceedings. He currently serves in the editorial board of Reliability Engineering and Systems Safety, Probabilistic Engineering Mechanics and Structural Safety. He promotes the dissemination of uncertainty quantification techniques through the development of the software UQLab and the community platform UQWorld.
