Saved in:
Bibliographic Details
Main Authors: Baez, Anthony, Zhang, Wang, Ma, Ziwen, Nguyen, Lam, Das, Subhro, Daniel, Luca
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.09048
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913158587744256
author Baez, Anthony
Zhang, Wang
Ma, Ziwen
Nguyen, Lam
Das, Subhro
Daniel, Luca
author_facet Baez, Anthony
Zhang, Wang
Ma, Ziwen
Nguyen, Lam
Das, Subhro
Daniel, Luca
contents We propose a novel projection method that guarantees the conservation of integral quantities in Physics-Informed Neural Networks (PINNs). While the soft constraint that PINNs use to enforce the structure of partial differential equations (PDEs) enables necessary flexibility during training, it also permits the discovered solution to violate physical laws. To address this, we introduce a projection method that guarantees the conservation of the linear and quadratic integrals, both separately and jointly. We derived the projection formulae by solving constrained non-linear optimization problems and found that our PINN modified with the projection, which we call PINN-Proj, reduced the error in the conservation of these quantities by three to four orders of magnitude compared to the soft constraint and marginally reduced the PDE solution error. We also found evidence that the projection improved convergence through improving the conditioning of the loss landscape. Our method holds promise as a general framework to guarantee the conservation of any integral quantity in a PINN if a tractable solution exists.
format Preprint
id arxiv_https___arxiv_org_abs_2511_09048
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Guaranteeing Conservation of Integrals with Projection in Physics-Informed Neural Networks
Baez, Anthony
Zhang, Wang
Ma, Ziwen
Nguyen, Lam
Das, Subhro
Daniel, Luca
Machine Learning
We propose a novel projection method that guarantees the conservation of integral quantities in Physics-Informed Neural Networks (PINNs). While the soft constraint that PINNs use to enforce the structure of partial differential equations (PDEs) enables necessary flexibility during training, it also permits the discovered solution to violate physical laws. To address this, we introduce a projection method that guarantees the conservation of the linear and quadratic integrals, both separately and jointly. We derived the projection formulae by solving constrained non-linear optimization problems and found that our PINN modified with the projection, which we call PINN-Proj, reduced the error in the conservation of these quantities by three to four orders of magnitude compared to the soft constraint and marginally reduced the PDE solution error. We also found evidence that the projection improved convergence through improving the conditioning of the loss landscape. Our method holds promise as a general framework to guarantee the conservation of any integral quantity in a PINN if a tractable solution exists.
title Guaranteeing Conservation of Integrals with Projection in Physics-Informed Neural Networks
topic Machine Learning
url https://arxiv.org/abs/2511.09048