Graph-based Prior and Forward Models for Inverse Problems on Manifolds with Boundaries

06/12/2021
by   John Harlim, et al.
0

This paper develops manifold learning techniques for the numerical solution of PDE-constrained Bayesian inverse problems on manifolds with boundaries. We introduce graphical Matérn-type Gaussian field priors that enable flexible modeling near the boundaries, representing boundary values by superposition of harmonic functions with appropriate Dirichlet boundary conditions. We also investigate the graph-based approximation of forward models from PDE parameters to observed quantities. In the construction of graph-based prior and forward models, we leverage the ghost point diffusion map algorithm to approximate second-order elliptic operators with classical boundary conditions. Numerical results validate our graph-based approach and demonstrate the need to design prior covariance models that account for boundary conditions.

READ FULL TEXT

page 9

page 18

page 21

page 22

06/07/2020

Ghost Point Diffusion Maps for solving elliptic PDE's on Manifolds with Classical Boundary Conditions

In this paper, we extend the class of kernel methods, the so-called diff...
04/10/2019

Unsupervised Deep Learning Algorithm for PDE-based Forward and Inverse Problems

We propose a neural network-based algorithm for solving forward and inve...
11/15/2019

Enforcing Boundary Conditions on Physical Fields in Bayesian Inversion

Inverse problems in computational mechanics consist of inferring physica...
10/23/2019

Kernel Methods for Bayesian Elliptic Inverse Problems on Manifolds

This paper investigates the formulation and implementation of Bayesian i...
02/12/2020

On parameter identification problems for elliptic boundary value problems in divergence form, Part I: An abstract framework

Parameter identification problems for partial differential equations are...
01/07/2020

MCMC for a hyperbolic Bayesian inverse problem in traffic flow modelling

As work on hyperbolic Bayesian inverse problems remains rare in the lite...
10/06/2020

Efficient Graph-based Tensile Strength Simulations of Random Fiber Structures

In this paper, we propose a model-simulation framework for virtual tensi...