A Perspective on Regression and Bayesian Approaches for System Identification of Pattern Formation Dynamics

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Wang, Zhenlin, Wu, Bowei, Garikipati, Krishna, Huan, Xun
Formato: Preprint
Publicado: 2020
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916178154225664
author Wang, Zhenlin
Wu, Bowei
Garikipati, Krishna
Huan, Xun
author_facet Wang, Zhenlin
Wu, Bowei
Garikipati, Krishna
Huan, Xun
contents We present two approaches to system identification, i.e. the identification of partial differential equations (PDEs) from measurement data. The first is a regression-based Variational System Identification procedure that is advantageous in not requiring repeated forward model solves and has good scalability to large number of differential operators. However it has strict data type requirements needing the ability to directly represent the operators through the available data. The second is a Bayesian inference framework highly valuable for providing uncertainty quantification, and flexible for accommodating sparse and noisy data that may also be indirect quantities of interest. However, it also requires repeated forward solutions of the PDE models which is expensive and hinders scalability. We provide illustrations of results on a model problem for pattern formation dynamics, and discuss merits of the presented methods.
format Preprint
id arxiv_https___arxiv_org_abs_2001_05646
institution arXiv
publishDate 2020
record_format arxiv
spellingShingle A Perspective on Regression and Bayesian Approaches for System Identification of Pattern Formation Dynamics
Wang, Zhenlin
Wu, Bowei
Garikipati, Krishna
Huan, Xun
Computational Physics
We present two approaches to system identification, i.e. the identification of partial differential equations (PDEs) from measurement data. The first is a regression-based Variational System Identification procedure that is advantageous in not requiring repeated forward model solves and has good scalability to large number of differential operators. However it has strict data type requirements needing the ability to directly represent the operators through the available data. The second is a Bayesian inference framework highly valuable for providing uncertainty quantification, and flexible for accommodating sparse and noisy data that may also be indirect quantities of interest. However, it also requires repeated forward solutions of the PDE models which is expensive and hinders scalability. We provide illustrations of results on a model problem for pattern formation dynamics, and discuss merits of the presented methods.
title A Perspective on Regression and Bayesian Approaches for System Identification of Pattern Formation Dynamics
topic Computational Physics
url https://arxiv.org/abs/2001.05646