Saved in:
Bibliographic Details
Main Authors: Yang, Weijie, Zhang, Xun, Jiang, Simin, Zhou, Yubao
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.13732
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914297687310336
author Yang, Weijie
Zhang, Xun
Jiang, Simin
Zhou, Yubao
author_facet Yang, Weijie
Zhang, Xun
Jiang, Simin
Zhou, Yubao
contents The estimation of high-dimensional physical parameters constrained by partial differential equations (PDEs) from limited and indirect measurements is a highly ill-posed problem. Traditional methods face significant accuracy and efficiency bottlenecks, particularly when observations are sparse, irregularly sampled, and constrained by real-world sensor placement. We propose the Physical Inversion Solver (PIS), a unified framework that couples Set-Conditioned Flow Matching with a Cosine-Annealed Sparsity Curriculum (CASC) to enable stable inversion from arbitrary, off-grid sensors even under minimal guidance. By leveraging straight-path transport, PIS achieves instantaneous inference (50 NFEs), offering orders-of-magnitude speedup over iterative baselines. Extensive experiments demonstrate that PIS reduces error by up to 88.7% under extreme sparsity (<1%) across subsurface characterization, wave-based characterization, and structural health monitoring, while providing robust uncertainty quantification for optimal sensor placement.
format Preprint
id arxiv_https___arxiv_org_abs_2512_13732
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PIS: A Generalized Physical Inversion Solver for Arbitrary Sparse Observations via Set Conditioned Flow Matching
Yang, Weijie
Zhang, Xun
Jiang, Simin
Zhou, Yubao
Machine Learning
Artificial Intelligence
The estimation of high-dimensional physical parameters constrained by partial differential equations (PDEs) from limited and indirect measurements is a highly ill-posed problem. Traditional methods face significant accuracy and efficiency bottlenecks, particularly when observations are sparse, irregularly sampled, and constrained by real-world sensor placement. We propose the Physical Inversion Solver (PIS), a unified framework that couples Set-Conditioned Flow Matching with a Cosine-Annealed Sparsity Curriculum (CASC) to enable stable inversion from arbitrary, off-grid sensors even under minimal guidance. By leveraging straight-path transport, PIS achieves instantaneous inference (50 NFEs), offering orders-of-magnitude speedup over iterative baselines. Extensive experiments demonstrate that PIS reduces error by up to 88.7% under extreme sparsity (<1%) across subsurface characterization, wave-based characterization, and structural health monitoring, while providing robust uncertainty quantification for optimal sensor placement.
title PIS: A Generalized Physical Inversion Solver for Arbitrary Sparse Observations via Set Conditioned Flow Matching
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2512.13732