Estimating systematic errors in Bayesian inversion using transport maps

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Casfor, Maren, Trunschke, Philipp, Heidenreich, Sebastian, Hegemann, Nando
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916958273798144
author Casfor, Maren
Trunschke, Philipp
Heidenreich, Sebastian
Hegemann, Nando
author_facet Casfor, Maren
Trunschke, Philipp
Heidenreich, Sebastian
Hegemann, Nando
contents In indirect measurements, the measurand is determined by solving an inverse problem which requires a model of the measurement process. Such models are often approximations and introduce systematic errors leading to a bias of the posterior distribution in Bayesian inversion. We propose a unified framework that combines transport maps from a reference distribution to the posterior distribution with the model error approach. This leads to an adaptive algorithm that jointly estimates the posterior distribution of the measurand and the model error. The efficiency and accuracy of the method are demonstrated on two model problems, showing that the approach effectively corrects biases while enabling fast sampling.
format Preprint
id arxiv_https___arxiv_org_abs_2509_16116
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Estimating systematic errors in Bayesian inversion using transport maps
Casfor, Maren
Trunschke, Philipp
Heidenreich, Sebastian
Hegemann, Nando
Methodology
Numerical Analysis
Probability
62G05, 62F15
In indirect measurements, the measurand is determined by solving an inverse problem which requires a model of the measurement process. Such models are often approximations and introduce systematic errors leading to a bias of the posterior distribution in Bayesian inversion. We propose a unified framework that combines transport maps from a reference distribution to the posterior distribution with the model error approach. This leads to an adaptive algorithm that jointly estimates the posterior distribution of the measurand and the model error. The efficiency and accuracy of the method are demonstrated on two model problems, showing that the approach effectively corrects biases while enabling fast sampling.
title Estimating systematic errors in Bayesian inversion using transport maps
topic Methodology
Numerical Analysis
Probability
62G05, 62F15
url https://arxiv.org/abs/2509.16116