Certified machine learning: Rigorous a posteriori error bounds for PDE defined PINNs

Abstract

Prediction error quantification in machine learning has been left out of most methodological investigations of neural networks, for both purely data-driven and physics-informed approaches. Beyond statistical investigations and generic results on the approximation capabilities of neural networks, we present a rigorous upper bound on the prediction error of physics-informed neural networks. This bound can be calculated without the knowledge of the true solution and only with a priori available information about the characteristics of the underlying dynamical system governed by a partial differential equation. We apply this a posteriori error bound exemplarily to four problems: the transport equation, the heat equation, the Navier-Stokes equation and the Klein-Gordon equation.

References in corpus (8)

• Gradient-enhanced physics-informed neural networks for forward and inverse PDE problems
• A Method for Representing Periodic Functions and Enforcing Exactly Periodic Boundary Conditions with Deep Neural Networks
• When Physics Meets Machine Learning: A Survey of Physics-Informed Machine Learning
• Generic bounds on the approximation error for physics-informed (and) operator learning
• Solving PDEs by Variational Physics-Informed Neural Networks: an a posteriori error analysis
• A Priori Analysis of Stable Neural Network Solutions to Numerical PDEs
• Certified machine learning: A posteriori error estimation for physics-informed neural networks
• Analysis of Deep Ritz Methods for Laplace Equations with Dirichlet Boundary Conditions