Bayesian Surrogate Analysis and Uncertainty Propagation with Explicit Surrogate Uncertainties and Implicit Spatio-temporal Correlations

01/11/2021
by   Sascha Ranftl, et al.
0

We introduce Bayesian Probability Theory to investigate uncertainty propagation based on meta-models. We approach the problem from the perspective of data analysis, with a given (however almost-arbitrary) input probability distribution and a given "training" set of computer simulations. While proven mathematically to be the unique consistent probability calculus, the subject of this paper is not to demonstrate beauty but usefulness. We explicitly list all propositions and lay open the general structure of any uncertainty propagation based on meta-models. The former allows rigorous treatment at any stage, while the latter allows us to quantify the interaction of the surrogate uncertainties with the usual parameter uncertainties. Additionally, we show a simple way to implicitly include spatio-temporal correlations. We then apply the framework jointly to a family of generalized linear meta-model that implicitly includes Polynomial Chaos Expansions as a special case. While we assume a Gaussian surrogate-uncertainty, we do not assume a scale for the surrogate uncertainty to be known, i.e. a Student-t. We end up with semi-analytic formulas for surrogate uncertainties and uncertainty propagation

READ FULL TEXT

page 1

page 2

page 3

page 4

research
06/13/2023

Epistemic and Aleatoric Uncertainty Quantification and Surrogate Modelling in High-Performance Multiscale Plasma Physics Simulations

This work suggests several methods of uncertainty treatment in multiscal...
research
05/31/2018

Uncertainty Quantification and Propagation of Imprecise Probabilities with Copula Dependence Modeling

Imprecise probability allows for partial probability specifications and ...
research
04/04/2020

A novel approach to bivariate meta-analysis of binary outcomes and its application in the context of surrogate endpoints

Bivariate meta-analysis provides a useful framework for combining inform...
research
01/31/2023

DiffSTG: Probabilistic Spatio-Temporal Graph Forecasting with Denoising Diffusion Models

Spatio-temporal graph neural networks (STGNN) have emerged as the domina...
research
05/10/2023

A Neural Emulator for Uncertainty Estimation of Fire Propagation

Wildfire propagation is a highly stochastic process where small changes ...
research
05/06/2019

Propagation of Uncertainties in Density-Driven Flow

Accurate modeling of contamination in subsurface flow and water aquifers...
research
08/08/2019

Spatial Flow-Field Approximation Using Few Thermodynamic Measurements Part II: Uncertainty Assessments

In this second part of our two-part paper, we provide a detailed, freque...

Please sign up or login with your details

Forgot password? Click here to reset