Lade Veranstaltungen

« Alle Veranstaltungen

  • Diese Veranstaltung hat bereits stattgefunden.

Felix-Klein-Kolloquium | Vortrag »Bayesian Inverse Problems – A Computational Perspective«

14. Januar 2020, 17:15 - 18:00

High- or infinite-dimensional inverse problems in the context of complex physical systems arise in many science and engineering applications, such as in parameter identification or data assimilation problems. Sampling-based Bayesian inference provides an approach that allows in principle to solve this problem and to provide a quantification of uncertainties without suffering from the curse of dimensionality. In contrast
to the very popular deep neural network approaches in machine learning, it comes with a theoretical framework that allows in many cases to rigorously analyse their performance and complexity. However, to bring these sampling methods in the feasible range for practical problems it is in general necessary to improve classical approaches, such as random-walk Metropolis-Hastings MCMC. I will discuss ways to directly accelerate classical Monte Carlo based inference approaches via multilevel or quasi-Monte Carlo ideas, as well as variational inference approaches that aim at finding a tractable transport map from a reference measure to the target measure for an efficient treatment of high dimensional Bayesian inverse problems.

Referent: Prof. Dr. Robert Scheichl, Universität Heidelberg

Der Vortrag findet um 17.15 Uhr im Raum 210 des Mathematik-Gebäudes 48 statt.

Studierende, Doktorandinnen und Doktoranden sowie Wissenschaftlerinnen und Wissenschaftler des Fachbereichs Mathematik der TU Kaiserslautern (TUK) und des Fraunhofer ITWM sind herzlich zum Kolloquium eingeladen!

Details

Datum:
14.01.2020
Zeit:
17:15 - 18:00
Veranstaltungskategorie:

Veranstaltungsort

TU Kaiserslautern, Geb. 48, Raum 210
Kaiserslautern, Deutschland