第 43 周 10 篇 覆盖区间: 2026-09-27 → 2026-10-20

第 43 周 · Norm-optimal model order reduction in reproducing kernel Hilbert spaces

第 43 周 Physics AI 论文速递,共 10 篇

本期要点

  • 本期 10 篇,聚焦 pde、物理信息约束
  • 推荐理由基于论文摘要自动生成,详见每条导读

本期收录

1

Norm-optimal model order reduction in reproducing kernel Hilbert spaces

Norm-optimal model order reduction in reproducing kernel Hilbert spaces
作者 Tamás Dózsa, Andrea Angino, Rolf Krause et al. 92 pages, 7 figures
为什么值得看

提出一种Norm-optimal model order reduction reproducing kernel的方法,在多个任务上验证了其有效性。

arXiv 原文 · 2610.28224 ↗ reduced-orderpdepartial differential equation
2

Low Residuals, Wrong Value: Learning the Viscosity Solution of Hamilton-Jacobi-Isaacs Equations with …

Low Residuals, Wrong Value: Learning the Viscosity Solution of Hamilton-Jacobi-Isaacs Equations with PINNs
作者 Raul Jimenez, Pablo Tejerina-Pérez, Leonid Sarieddine et al. 43 pages, 9 figures
为什么值得看

提出一种Residuals, Wrong Value: Viscosity Solution Hamilton-Jacobi-Isaacs的方法,在多个任务上验证了其有效性。

arXiv 原文 · 2610.27193 ↗ physics-informedpinnpinnspde
3

Synthetic Dataset Generation, Physical Consistency and Active Learning for Shape-Informed Surrogate Modeling …

Synthetic Dataset Generation, Physical Consistency and Active Learning for Shape-Informed Surrogate Modeling of Hemodynamics
作者 Francesco Romor, Richard Schussnig, Charles A. Taylor et al.
为什么值得看

提出一种Synthetic Dataset Generation, Physical Consistency Active的方法,在多个任务上验证了其有效性。

arXiv 原文 · 2610.27130 ↗ physics-informedsurrogate modelcomputational fluid dynamics
4

A mixed leading variable physics-informed neural network for nonsmooth dynamics simulation in …

A mixed leading variable physics-informed neural network for nonsmooth dynamics simulation in contact systems
作者 Xinyuan Li, Zilin Li, Hongtao Wei et al. 41 pages, 17 figures
为什么值得看

提出一种mixed leading variable physics-informed network nonsmooth的方法,在多个任务上验证了其有效性。

5

PINN-LPCDD: Physics-Informed Neural Network with Learnable and Physically Consistent Domain Decomposition for …

PINN-LPCDD: Physics-Informed Neural Network with Learnable and Physically Consistent Domain Decomposition for Compressible Multi-Medium Flows
作者 Weiheng Zeng, Hao Wang, Jiaxin Zhu et al.
为什么值得看

提出一种PINN-LPCDD: Physics-Informed Network Learnable Physically Consistent的方法,在多个任务上验证了其有效性。

arXiv 原文 · 2610.22903 ↗ physics-informedpinnpinns
6

Learnable Splines for Unstructured Neural Operators

Learnable Splines for Unstructured Neural Operators
作者 Vignesh Gopakumar, Zhuoyuan Wang, Ander Gray et al.
为什么值得看

探讨了Learnable Splines Unstructured Operators的相关问题,值得关注。

arXiv 原文 · 2610.22446 ↗ neural operatorfourier neuralpartial differential equation
7

Online reduced-order modeling from primal solver trajectories for PDE-constrained optimization

Online reduced-order modeling from primal solver trajectories for PDE-constrained optimization
作者 Zhe Lu, Wenbo Cao, Weiwei Zhang 25 pages
为什么值得看

探索了Online reduced-order modeling primal solver trajectories在实际场景中的应用效果。

8

SBNO : Schrödinger Bridge Neural Operator for the Forward and Inverse Problems …

SBNO : Schrödinger Bridge Neural Operator for the Forward and Inverse Problems under Missing Information
作者 Jaehyeon Park, Inhyeok Jeong, Noseong Park NeurIPS 2026
为什么值得看

提出一种SBNO Schrödinger Bridge Operator Forward Inverse的方法,在多个任务上验证了其有效性。

arXiv 原文 · 2610.13317 ↗ surrogate modelpdepartial differential equation
9

SINO: Scale-Invariant Neural Operator

SINO: Scale-Invariant Neural Operator
作者 Kaichen Ouyang, Chenglei Yu, Chuanrui Wang et al. 38 pages, 9 figures
为什么值得看

提出一种SINO: Scale-Invariant Operator的方法,在多个任务上验证了其有效性。

arXiv 原文 · 2609.36890 ↗ neural operatordeeponetscientific machine learningpartial differential equation
10

PI-NOMT: Physics-Informed Neural Optimal Mass Transport for Brain Fluid Dynamics

PI-NOMT: Physics-Informed Neural Optimal Mass Transport for Brain Fluid Dynamics
作者 Mehmet Emin Acar, Vahit Bugra Yesilkaynak, Helene Benveniste et al.
为什么值得看

探讨了PI-NOMT: Physics-Informed Optimal Mass Transport Brain的相关问题,值得关注。

arXiv 原文 · 2609.33857 ↗ physics-informedscientific machine learningpde

本期正文

本期导读

第 43 周 Physics AI 论文速递,共 10 篇

本期要点

  • 本期 10 篇,聚焦 pde、物理信息约束
  • 推荐理由基于论文摘要自动生成,详见每条导读

论文推荐

1. Norm-optimal model order reduction in reproducing kernel Hilbert spaces

提出一种Norm-optimal model order reduction reproducing kernel的方法,在多个任务上验证了其有效性。

来源:92 pages, 7 figures

2. Low Residuals, Wrong Value: Learning the Viscosity Solution of Hamilton-Jacobi-Isaacs Equations with …

提出一种Residuals, Wrong Value: Viscosity Solution Hamilton-Jacobi-Isaacs的方法,在多个任务上验证了其有效性。

来源:43 pages, 9 figures

3. Synthetic Dataset Generation, Physical Consistency and Active Learning for Shape-Informed Surrogate Modeling …

提出一种Synthetic Dataset Generation, Physical Consistency Active的方法,在多个任务上验证了其有效性。

4. A mixed leading variable physics-informed neural network for nonsmooth dynamics simulation in …

提出一种mixed leading variable physics-informed network nonsmooth的方法,在多个任务上验证了其有效性。

来源:41 pages, 17 figures

5. PINN-LPCDD: Physics-Informed Neural Network with Learnable and Physically Consistent Domain Decomposition for …

提出一种PINN-LPCDD: Physics-Informed Network Learnable Physically Consistent的方法,在多个任务上验证了其有效性。

6. Learnable Splines for Unstructured Neural Operators

探讨了Learnable Splines Unstructured Operators的相关问题,值得关注。

7. Online reduced-order modeling from primal solver trajectories for PDE-constrained optimization

探索了Online reduced-order modeling primal solver trajectories在实际场景中的应用效果。

来源:25 pages

8. SBNO : Schrödinger Bridge Neural Operator for the Forward and Inverse Problems …

提出一种SBNO Schrödinger Bridge Operator Forward Inverse的方法,在多个任务上验证了其有效性。

来源:NeurIPS 2026

9. SINO: Scale-Invariant Neural Operator

提出一种SINO: Scale-Invariant Operator的方法,在多个任务上验证了其有效性。

来源:38 pages, 9 figures

10. PI-NOMT: Physics-Informed Neural Optimal Mass Transport for Brain Fluid Dynamics

探讨了PI-NOMT: Physics-Informed Optimal Mass Transport Brain的相关问题,值得关注。


数据来自 arXiv API(scripts/fetch-arxiv-digest.mjs),仅收录元数据与站点自撰推荐理由,不转载摘要全文。 收录标准:目标分类(cs.LG / physics.flu-dyn / physics.comp-ph / math.NA)× 核心词相关度 ≥ 2 × 最近 45 天提交 × 与站内已收录论文不重复。