Quantifying uncertainty in the numerical integration of evolution equations based on Bayesian isotonic regression

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Miyatake, Yuto, Irie, Kaoru, Matsuda, Takeru
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917835897307136
author Miyatake, Yuto
Irie, Kaoru
Matsuda, Takeru
author_facet Miyatake, Yuto
Irie, Kaoru
Matsuda, Takeru
contents This paper presents a new Bayesian framework for quantifying discretization errors in numerical solutions of ordinary differential equations. By modelling the errors as random variables, we impose a monotonicity constraint on the variances, referred to as discretization error variances. The key to our approach is the use of a shrinkage prior for the variances coupled with variable transformations. This methodology extends existing Bayesian isotonic regression techniques to tackle the challenge of estimating the variances of a normal distribution. An additional key feature is the use of a Gaussian mixture model for the $\log$-$χ^2_1$ distribution, enabling the development of an efficient Gibbs sampling algorithm for the corresponding posterior.
format Preprint
id arxiv_https___arxiv_org_abs_2411_08338
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Quantifying uncertainty in the numerical integration of evolution equations based on Bayesian isotonic regression
Miyatake, Yuto
Irie, Kaoru
Matsuda, Takeru
Numerical Analysis
Data Analysis, Statistics and Probability
Methodology
This paper presents a new Bayesian framework for quantifying discretization errors in numerical solutions of ordinary differential equations. By modelling the errors as random variables, we impose a monotonicity constraint on the variances, referred to as discretization error variances. The key to our approach is the use of a shrinkage prior for the variances coupled with variable transformations. This methodology extends existing Bayesian isotonic regression techniques to tackle the challenge of estimating the variances of a normal distribution. An additional key feature is the use of a Gaussian mixture model for the $\log$-$χ^2_1$ distribution, enabling the development of an efficient Gibbs sampling algorithm for the corresponding posterior.
title Quantifying uncertainty in the numerical integration of evolution equations based on Bayesian isotonic regression
topic Numerical Analysis
Data Analysis, Statistics and Probability
Methodology
url https://arxiv.org/abs/2411.08338