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

10/07/2022
by   Birgit Hillebrecht, et al.
0

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.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/31/2022

Certified machine learning: A posteriori error estimation for physics-informed neural networks

Physics-informed neural networks (PINNs) are one popular approach to int...
research
08/23/2019

Physics Informed Data Driven model for Flood Prediction: Application of Deep Learning in prediction of urban flood development

Flash floods in urban areas occur with increasing frequency. Detecting t...
research
10/18/2020

An energy-based error bound of physics-informed neural network solutions in elasticity

An energy-based a posteriori error bound is proposed for the physics-inf...
research
07/14/2023

Inverse Evolution Layers: Physics-informed Regularizers for Deep Neural Networks

This paper proposes a novel approach to integrating partial differential...
research
12/14/2022

Harmonic (Quantum) Neural Networks

Harmonic functions are abundant in nature, appearing in limiting cases o...
research
04/07/2023

EPINN-NSE: Enhanced Physics-Informed Neural Networks for Solving Navier-Stokes Equations

Fluid mechanics is a fundamental field in engineering and science. Solvi...
research
02/26/2023

Efficient physics-informed neural networks using hash encoding

Physics-informed neural networks (PINNs) have attracted a lot of attenti...

Please sign up or login with your details

Forgot password? Click here to reset