Machine Learning Conservation Laws of Dynamical systems
Computational Physics
2024-06-03 v1
Abstract
Conservation laws are of great theoretical and practical interest. We describe a novel approach to machine learning conservation laws of finite-dimensional dynamical systems using trajectory data. It is the first such approach based on kernel methods instead of neural networks which leads to lower computational costs and requires a lower amount of training data. We propose the use of an "indeterminate" form of kernel ridge regression where the labels still have to be found by additional conditions. We use here a simple approach minimising the length of the coefficient vector to discover a single conservation law.
Cite
@article{arxiv.2405.20857,
title = {Machine Learning Conservation Laws of Dynamical systems},
author = {Meskerem Abebaw Mebratie and Rüdiger Nather and Guido Falk von Rudorff and Werner M. Seiler},
journal= {arXiv preprint arXiv:2405.20857},
year = {2024}
}