Saved in:
Bibliographic Details
Main Authors: Mukherjee, Amartya, Stadt, Melissa M., Podina, Lena, Kohandel, Mohammad, Liu, Jun
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.08563
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913234909396992
author Mukherjee, Amartya
Stadt, Melissa M.
Podina, Lena
Kohandel, Mohammad
Liu, Jun
author_facet Mukherjee, Amartya
Stadt, Melissa M.
Podina, Lena
Kohandel, Mohammad
Liu, Jun
contents Diffusion models have emerged as a promising class of generative models that map noisy inputs to realistic images. More recently, they have been employed to generate solutions to partial differential equations (PDEs). However, they still struggle with inverse problems in the Laplacian operator, for instance, the Poisson equation, because the eigenvalues that are large in magnitude amplify the measurement noise. This paper presents a novel approach for the inverse and forward solution of PDEs through the use of denoising diffusion restoration models (DDRM). DDRMs were used in linear inverse problems to restore original clean signals by exploiting the singular value decomposition (SVD) of the linear operator. Equivalently, we present an approach to restore the solution and the parameters in the Poisson equation by exploiting the eigenvalues and the eigenfunctions of the Laplacian operator. Our results show that using denoising diffusion restoration significantly improves the estimation of the solution and parameters. Our research, as a result, pioneers the integration of diffusion models with the principles of underlying physics to solve PDEs.
format Preprint
id arxiv_https___arxiv_org_abs_2402_08563
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Denoising Diffusion Restoration Tackles Forward and Inverse Problems for the Laplace Operator
Mukherjee, Amartya
Stadt, Melissa M.
Podina, Lena
Kohandel, Mohammad
Liu, Jun
Machine Learning
Computer Vision and Pattern Recognition
Analysis of PDEs
Diffusion models have emerged as a promising class of generative models that map noisy inputs to realistic images. More recently, they have been employed to generate solutions to partial differential equations (PDEs). However, they still struggle with inverse problems in the Laplacian operator, for instance, the Poisson equation, because the eigenvalues that are large in magnitude amplify the measurement noise. This paper presents a novel approach for the inverse and forward solution of PDEs through the use of denoising diffusion restoration models (DDRM). DDRMs were used in linear inverse problems to restore original clean signals by exploiting the singular value decomposition (SVD) of the linear operator. Equivalently, we present an approach to restore the solution and the parameters in the Poisson equation by exploiting the eigenvalues and the eigenfunctions of the Laplacian operator. Our results show that using denoising diffusion restoration significantly improves the estimation of the solution and parameters. Our research, as a result, pioneers the integration of diffusion models with the principles of underlying physics to solve PDEs.
title Denoising Diffusion Restoration Tackles Forward and Inverse Problems for the Laplace Operator
topic Machine Learning
Computer Vision and Pattern Recognition
Analysis of PDEs
url https://arxiv.org/abs/2402.08563