26–27 Feb 2026
University of Graz
Europe/Vienna timezone

Certifying physics-informed neural networks through lower error bounds

26 Feb 2026, 15:24
9m
HS11.02 (University of Graz)

HS11.02

University of Graz

Department of Mathematics and Scientific Computing Heinrichstraße 36 8010 Graz
Poster Presentation

Speaker

Arzu Ahmadova (University of Duisburg-Essen)

Description

Physics-informed neural networks (PINNs) bring together machine learning and physical laws to solve differential equations. While Hillebrecht and Unger (2022) provide rigorous a posteriori upper bounds for PINN prediction errors, certification requires complementary lower bounds to establish complete error enclosures. In this paper, we derive computable a posteriori lower bounds for PINN errors in ODEs under strong monotonicity conditions. These bounds rely solely on the neural network approximation and the ODE residual, requiring no a priori knowledge of the true solution. This work gives fully certified error bands for nonlinear ODEs and for linear ODEs satisfying structural assumptions, providing robust bounds without needing a lot of training data.

Affiliation

University of Duisburg-Essen

Author

Arzu Ahmadova (University of Duisburg-Essen)

Presentation materials

There are no materials yet.