IRTG Modern Inverse Problems (MIP)
Lade Veranstaltungen

« Alle Veranstaltungen

  • Diese Veranstaltung hat bereits stattgefunden.

SSD Seminar Series with Prof. Dr. Anne K. Reinarz

3. Mai 2021 @ 16:00 - 17:00

Prof. Dr. Anne K. Reinarz – Bayesian Inverse Problems for Tsunami Simulation

Department of Computer Science Durham University, UK



The ExaHyPE Engine is a hyperbolic PDE engine capable of solving systems of first order hyperbolic PDEs. The engine provides a space-tree discretization of the computational domain, higher-order DG schemes and a-posteriori subcell limiters. In this talk I will introduce ExaHyPE, focusing on a solver for the Shallow Water equations (SWE). This solver uses a well-balanced numerical scheme and can handle wetting and drying making it suitable for the simulation of tsunamis.

ExaHyPE was initially designed only with forward modeling in mind. In order to utilise expertise from the uncertainty quantification (UQ) community we have extended it with an interface to the MIT UQ library (MUQ). I will present a parallelization strategy for multilevel Markov chain Monte Carlo, a state-of-the-art, algorithmically scalable UQ algorithm for Bayesian inverse problems. The integration between MUQ and ExaHyPE allows for large-scale parallelism across forward model evaluations and the UQ algorithms themselves. The main scalability challenge presents itself in the form of strong data dependencies introduced by the MLMCMC method, prohibiting trivial parallelization. I will demonstrate the method by using it to infer the most probable locations for the initialization of a tsunami from buoy data.


3. Mai 2021
16:00 - 17:00


a link for the Zoom meeting room will be send in the newsletter one week before the seminar starts. If you need any organizational help please contact office@aices.rwth-aachen.de