Saved in:
Bibliographic Details
Main Authors: Cao, Xiang, Zhang, Xiaoqun
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2408.17333
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909367385718784
author Cao, Xiang
Zhang, Xiaoqun
author_facet Cao, Xiang
Zhang, Xiaoqun
contents Diffusion models have been widely studied as effective generative tools for solving inverse problems. The main ideas focus on performing the reverse sampling process conditioned on noisy measurements, using well-established numerical solvers for gradient updates. Although diffusion-based sampling methods can produce high-quality reconstructions, challenges persist in nonlinear PDE-based inverse problems and sampling speed. In this work, we explore solving PDE-based travel-time tomography based on subspace diffusion generative models. Our main contributions are twofold: First, we propose a posterior sampling process for PDE-based inverse problems by solving the associated adjoint-state equation. Second, we resorted to the subspace-based dimension reduction technique for conditional sampling acceleration, enabling solving the PDE-based inverse problems from coarse to refined grids. Our numerical experiments showed satisfactory advancements in improving the travel-time imaging quality and reducing the sampling time for reconstruction.
format Preprint
id arxiv_https___arxiv_org_abs_2408_17333
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Subspace Diffusion Posterior Sampling for Travel-Time Tomography
Cao, Xiang
Zhang, Xiaoqun
Numerical Analysis
Analysis of PDEs
Diffusion models have been widely studied as effective generative tools for solving inverse problems. The main ideas focus on performing the reverse sampling process conditioned on noisy measurements, using well-established numerical solvers for gradient updates. Although diffusion-based sampling methods can produce high-quality reconstructions, challenges persist in nonlinear PDE-based inverse problems and sampling speed. In this work, we explore solving PDE-based travel-time tomography based on subspace diffusion generative models. Our main contributions are twofold: First, we propose a posterior sampling process for PDE-based inverse problems by solving the associated adjoint-state equation. Second, we resorted to the subspace-based dimension reduction technique for conditional sampling acceleration, enabling solving the PDE-based inverse problems from coarse to refined grids. Our numerical experiments showed satisfactory advancements in improving the travel-time imaging quality and reducing the sampling time for reconstruction.
title Subspace Diffusion Posterior Sampling for Travel-Time Tomography
topic Numerical Analysis
Analysis of PDEs
url https://arxiv.org/abs/2408.17333