Learning of discrete models of variational PDEs from data

Abstract

We show how to learn discrete field theories from observational data of fields on a space-time lattice. For this, we train a neural network model of a discrete Lagrangian density such that the discrete Euler--Lagrange equations are consistent with the given training data. We, thus, obtain a structure-preserving machine learning architecture. Lagrangian densities are not uniquely defined by the solutions of a field theory. We introduce a technique to derive regularisers for the training process which optimise numerical regularity of the discrete field theory. Minimisation of the regularisers guarantees that close to the training data the discrete field theory behaves robust and efficient when used in numerical simulations. Further, we show how to identify structurally simple solutions of the underlying continuous field theory such as travelling waves. This is possible even when travelling waves are not present in the training data. This is compared to data-driven model order reduction based approaches, which struggle to identify suitable latent spaces containing structurally simple solutions when these are not present in the training data. Ideas are demonstrated on examples based on the wave equation and the Schrödinger equation.

References in corpus (6)

• On singular Lagrangian underlying the Schrödinger equation
• Variational Learning of Euler-Lagrange Dynamics from Data
• Preserving Lagrangian structure in data-driven reduced-order modeling of large-scale dynamical systems
• Hamiltonian Neural Networks with Automatic Symmetry Detection
• Discrete Lagrangian Neural Networks with Automatic Symmetry Discovery
• Backward error analysis for conjugate symplectic methods

Cited by in corpus (2)

• Machine learning of continuous and discrete variational ODEs with convergence guarantee and uncertainty quantification
• Lagrange top: integrability according to Liouville and examples of analytic solutions