CMCC Webinar
05 March 2024, 11:00 – 13:00 CET
To join the webinar, register here
Speaker
Jonathan Lindbloom, Darmouth College
Abstract
Bayesian hierarchical models have been demonstrated to provide efficient algorithms for finding sparse solutions to ill-posed inverse problems. The models typically employ a conditionally Gaussian prior model for the unknown source, augmented by a hyper-prior governing auxiliary local variance parameters. In this talk, we review the methodology and present our contributions towards (1) generalizing the applicability of these models, (2) generalizing their convexity and convergence analysis, and (3) the development of efficient, scalable computational procedures for characterizing their posterior distributions. In particular, we discuss both MAP estimation and sample-based uncertainty quantification. We illustrate the performance of our methods with a suite of numerical examples and applications.
WATCH THE VIDEO
HOW TO PARTICIPATE
05 March 2024, 11:00 – 13:00 CET
To join the webinar, register here
ORGANIZED BY:
CMCC