An Unconditional Representation of the Conditional Score in Infinite-Dimensional Linear Inverse Problems
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914532703600640 |
|---|---|
| author | Schneider, Fabian Duong, Duc-Lam Lassas, Matti de Hoop, Maarten V. Helin, Tapio |
| author_facet | Schneider, Fabian Duong, Duc-Lam Lassas, Matti de Hoop, Maarten V. Helin, Tapio |
| contents | Score-based diffusion models (SDMs) have emerged as a powerful tool for sampling from the posterior distribution in Bayesian inverse problems. However, existing methods often require multiple evaluations of the forward mapping to generate a single sample, resulting in significant computational costs for large-scale inverse problems. To address this, we propose an unconditional representation of the conditional score function (UCoS) tailored to linear inverse problems, which avoids forward model evaluations during sampling by shifting computational effort to an offline training phase. In this phase, a \emph{task-dependent} score function is learned based on the linear forward operator. Crucially, we show that the conditional score can be derived \emph{exactly} from a trained (unconditional) score using affine transformations, eliminating the need for conditional score approximations. Our approach is formulated in infinite-dimensional function spaces, making it inherently discretization-invariant. We support this formulation with a rigorous convergence analysis that justifies UCoS beyond any specific discretization. Finally we validate UCoS through high-dimensional computed tomography (CT) and image deblurring experiments, demonstrating both scalability and accuracy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_15643 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | An Unconditional Representation of the Conditional Score in Infinite-Dimensional Linear Inverse Problems Schneider, Fabian Duong, Duc-Lam Lassas, Matti de Hoop, Maarten V. Helin, Tapio Machine Learning Numerical Analysis Analysis of PDEs Probability 62F15, 65N21, 68Q32, 60Hxx, 60Jxx, 68T07, 92C55 Score-based diffusion models (SDMs) have emerged as a powerful tool for sampling from the posterior distribution in Bayesian inverse problems. However, existing methods often require multiple evaluations of the forward mapping to generate a single sample, resulting in significant computational costs for large-scale inverse problems. To address this, we propose an unconditional representation of the conditional score function (UCoS) tailored to linear inverse problems, which avoids forward model evaluations during sampling by shifting computational effort to an offline training phase. In this phase, a \emph{task-dependent} score function is learned based on the linear forward operator. Crucially, we show that the conditional score can be derived \emph{exactly} from a trained (unconditional) score using affine transformations, eliminating the need for conditional score approximations. Our approach is formulated in infinite-dimensional function spaces, making it inherently discretization-invariant. We support this formulation with a rigorous convergence analysis that justifies UCoS beyond any specific discretization. Finally we validate UCoS through high-dimensional computed tomography (CT) and image deblurring experiments, demonstrating both scalability and accuracy. |
| title | An Unconditional Representation of the Conditional Score in Infinite-Dimensional Linear Inverse Problems |
| topic | Machine Learning Numerical Analysis Analysis of PDEs Probability 62F15, 65N21, 68Q32, 60Hxx, 60Jxx, 68T07, 92C55 |
| url | https://arxiv.org/abs/2405.15643 |