Domain decomposition and uncertainty quantification for large scale wave problems

In the first part of the talk, I will discuss domain decomposition methods for large scale time-harmonic flow acoustic problems, with a focus on non-reflecting boundary conditions and their application as transmission operator for domain decomposition. I will also present industrial benchmark results in a high-performance computing context. In the second part of the talk, I will address uncertainty quantification in Full Waveform Inversion (FWI). Revisiting the classical minimization scheme for FWI, I will explain how the theory of gradient flows can inspire novel algorithms to quantify uncertainties.