Physics-Informed Neural Embeddings of PDE Solution Families

DOI10.48550/arXiv.2607.06348MaRDI QIDQ7283944arXivFDO

Authors Svitlana Mayboroda, Pavlos Protopapas, Leonid Sarieddine, David N. Spergel, Pedro Tarancón-Álvarez, Raul Jimenez

Publication date 7 July 2026

Published in arXiv

Tldr Physics-Informed Neural Network (PINN) learns embeddings of PDE solution families by leveraging a shared body to learn a latent manifold representing the solution space and linear heads reconstructing individual solutions with different initial conditions; orthogonalization penalty removes degeneracies, allowing for dimensionality reduction and reproducible spectral analysis.

Full work available at URL https://doi.org/10.48550/arXiv.2607.06348

arXiv classification

cs.LGmath.NAphysics.comp-ph

Cites work

• Applied Analysis of the Navier-Stokes Equations
• B-PINNs: Bayesian physics-informed neural networks for forward and inverse PDE problems with noisy data
• Deep learning
• DeepXDE: a deep learning library for solving differential equations
• DGM: a deep learning algorithm for solving partial differential equations
• Dynamic mode decomposition of numerical and experimental data
• Efficient PINNs: Multi-Head Unimodular Regularization of the Solutions Space
• Fourier Neural Operator for Parametric Partial Differential Equations
• Fourier Neural Operator with Learned Deformations for PDEs on General Geometries
• Generalized reduced basis methods and n-width estimates for the approximation of the solution manifold of parametric PDEs
• Gradient-annihilated PINNs for solving Riemann problems: application to relativistic hydrodynamics
• Gravitational duals from equations of state
• Neural‐network‐based approximations for solving partial differential equations
• Nonuniqueness of weak solutions to the Navier-Stokes equation
• One-Shot Transfer Learning for Nonlinear PDEs with Perturbative PINNs
• Physics-Informed Deep Neural Operator Networks
• Reduced basis methods for partial differential equations. An introduction
• Spectral properties of dynamical systems, model reduction and decompositions
• Structure-preserving algorithms for the two-dimensional sine-Gordon equation with Neumann boundary conditions
• The Proper Orthogonal Decomposition in the Analysis of Turbulent Flows

This page was built for publication: Physics-Informed Neural Embeddings of PDE Solution Families

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7283944)