When Data Driven Reduced Order Modeling Meets Full Waveform Inversion - Rothschild Lecture

Explore a comprehensive lecture on data-driven reduced order modeling and its application to full waveform inversion in this Rothschild Lecture by Professor Liliana Borcea from the University of Michigan. Delve into the inverse problem for the wave equation, focusing on determining variable wave speed from backscattered wave data collected by sensors. Examine the two main types of methods in inverse backscattering: qualitative imaging methods and quantitative velocity estimation techniques. Discover the challenges of PDE constrained optimization in velocity estimation, particularly for high-frequency data. Learn about a novel approach using a data-driven reduced order model (ROM) of the wave operator, which offers efficient computation and improved objective function definition for velocity estimation. Gain insights into this interdisciplinary field with applications in geophysical exploration, radar imaging, and non-destructive material evaluation.