Mitigating numerical stiffness in least-squares formulations of elliptic PDEs for physics-informed neural networks
Machine Learning: Science and Technology · 09.11.2026
DOI: 10.1088/2632-2153/aebff5 ↗Kurz gefasst
Fachartikel in der Zeitschrift Machine Learning: Science and Technology, online erschienen am 09.11.2026. Verfasst von Phil-Alexander Hofmann und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Saxon Ministry for Science, Culture and Tourism, Germany's Federal Ministry of Education and Research, Center for Advanced Systems Understanding.
Abstract (Originalsprache)
Residual-based physics-informed neural networks (PINNs) combine interior PDE and boundary residuals using mean squared errors. 
For elliptic boundary value problems, these formulations can show numerical stiffness causing slow training.
We study a residual formulation that combines an H 0 −1 interior residual with an H 1/2 boundary residual for nonzero Dirichlet boundary data.
For finite-dimensional test spaces in H 1 (Ω), we introduce two finite-dimensional approximations, the subspace-induced negative-order Sobolev seminorm (SINOS) and its dual counterpart (D-SINOS), and quantify their stiffness through a stiffness ratio defined with respect to the H 1 -norm.
We establish continuity and coercivity estimates for both formulations and show that, for polynomial spaces of degree n, the stiffness ratios of SINOS and D-SINOS grow at most as O(n 2 ) and O(1), respectively.
Numerical experiments recover rates of n 2.00 and n 0.05 , compared with n 6.95 and n 5.68 for MSE and L 2 -based residual losses.
PINN experiments for 
the Poisson equation, 
a stationary convection-diffusion-reaction problem, 
and the stationary incompressible Navier-Stokes equations 
show that SINOS achieves consistently strong convergence and accuracy, while D-SINOS improves over MSE but does not attain the performance of SINOS.
These results demonstrate that negative-order Sobolev residuals substantially reduce the stiffness of least-squares PINN formulations. 
At the same time, the superior PINN performance of SINOS despite the smaller stiffness ratio of D-SINOS demonstrates that stiffness mitigation alone does not determine optimization and approximation quality.
Förderung
- Saxon Ministry for Science, Culture and Tourism
- Germany's Federal Ministry of Education and Research
- Center for Advanced Systems Understanding
- Saxon State Parliament
Zitieren
APA
Hofmann, P., & Hecht, M. (2026). Mitigating numerical stiffness in least-squares formulations of elliptic PDEs for physics-informed neural networks. Machine Learning: Science and Technology. https://doi.org/10.1088/2632-2153/aebff5
Vancouver
Hofmann P, Hecht M. Mitigating numerical stiffness in least-squares formulations of elliptic PDEs for physics-informed neural networks. Machine Learning: Science and Technology. 2026. doi:10.1088/2632-2153/aebff5
BibTeX
@article{Hofmann2026,
author = {Hofmann, Phil-Alexander and Hecht, Michael},
title = {Mitigating numerical stiffness in least-squares formulations of elliptic PDEs for physics-informed neural networks},
journal = {Machine Learning: Science and Technology},
year = {2026},
doi = {10.1088/2632-2153/aebff5}
}
EUVIMED