Low-rank modelling in uncertainty quantification

Mini Symposium

Uncertainties in modelling and simulation of dynamical systems come from different sources: modelled or unmodelled external perturbations, unknown or stochastically changing model parameters, and measurement errors, to name a few. In any case, respecting these uncertainties in the model equations and their discretization means adding yet another model dimension. Thus, since the computational complexity grows exponentially with the dimensions, low-rank descriptions seem to be a promising way to arrive at feasible schemes for models that include PDEs. This mini symposium features suitably extended common approaches as well as newly developed methods that have the identification and exploitation of low-rank structures as the common denominator.

Session 1

Title
Karsten Urban, Uni Ulm Parameter Functions within Model Reduction for Uncertainty Quantification
Martin Grepl, RWTH Aachen A certified reduced basis approach to variational data assimilation
Yoshihito Kazashi, EPF Lausanne Analysis of the dynamical low rank equations for random semi-linear evolutionary problems
Alexandra Bünger, TU Chemnitz Low-rank methods and iso-geometric analysis

Session 2

Title
Martin Redmann, WIAS Berlin Energy estimates and model order reduction for bilinear systems with Lévy noise
Yue Qiu, Max Planck Institute for Dynamics of Complex Technical Systems Low-rank ensemble Kalman filter for nonlinear networks: a gas network example
Ralf Zimmermann, SDU Odense Low-rank parameterizations for the unsteady Navier-stokes equations in the frequency domain
Martin Hess, SISSA Triest Model Reduction in Micromotility Applications
Avatar
Jan Heiland
Team leader and Jun.-Prof. in Applied Mathematics

My research interests include control systems, differential-algebraic equations, and flow problems.