Speaker
Description
To leverage the full potential of next-generation gravitational wave observatories, such as the Einstein Telescope (in Europe) and LISA (in space), we will need to improve the efficiency of numerical relativity (NR) simulations in terms of computational speed and accuracy. In this talk, we will introduce probabilistic numerics as a promising new approach to gain control over uncertainties in NR. Instead of a single estimate, probabilistic numerical methods model a full probability distribution of the solution, allowing to quantify the uncertainty associated with the model parameters and numerical discretization schemes. This uncertainty quantification can then be used to optimize simulations. This talk will provide a first demonstration of this approach with a probabilistic numerical solver for the Einstein equations.
