Speaker
Arved Bartuska
Description
Measuring the expected information gain (EIG) of an experiment allows for comprehensive design optimization. Efficiently estimating the EIG is crucial when data are scarce or costly to obtain. We propose several estimators combining the randomized quasi-Monte Carlo method with Laplace-based importance sampling, and showcase their efficiency both theoretically and via numerical examples.
Primary author
Arved Bartuska
Co-authors
Dr
André Carlon
(Computer, Electrical and Mathematical Sciences and Engineering, KAUST)
Luis Espath
(Faculty of Science, University of Nottingham, United Kingdom)
Raúl Tempone
(Computer, Electrical and Mathematical Sciences and Engineering, KAUST, and Alexander von Humboldt professor in Mathematics of Uncertainty Quantification, RWTH Aachen University)
Sebastian Krumscheid
(Steinbuch Center for Computing and Institute for Applied and Numerical Mathematics, Karlsruhe Institute of Technology)