Explore deep Gaussian processes for Bayesian inversion in this 26-minute conference talk by Matt Dunlop from Courant. Delve into uncertainty quantification techniques for better understanding physical systems and decision-making under uncertainty. Learn how Gaussian Process emulators can replace complex, computationally expensive codes for more efficient modeling. Examine the theoretical and numerical aspects of GP emulation, with a focus on applications to large-scale problems in climate, tsunami, and earthquake research. Cover key topics including Bayesian inversion, deep Gaussian processes, composition-based processes, methods, numerical examples, and future directions in this field.
Deep Gaussian Processes for Bayesian Inversion - Matt Dunlop, Courant
Alan Turing Institute via YouTube
Overview
Syllabus
Introduction
Bayesian Inversion
Deep Gaussian Processes
Composition Based Processes
Methods
Numerical Examples
Future Directions
Taught by
Alan Turing Institute
Related Courses
-
Deep and Multi-Fidelity Learning with Gaussian Processes
-
Introduction to Uncertainty Quantification
-
Introduction to Scientific Machine Learning
-
Some Thoughts on Gaussian Processes for Emulation of Deterministic Computer Models
-
Known Boundary Emulation of Complex Computer Models - Ian Vernon, Cambridge
-
High-Dimensional Hierarchical Models for Large-Scale Geophysical Applications
