Improving Simulations with Symmetry Control Neural Networks
Machine Learning
2021-04-30 v1 Computational Physics
Abstract
The dynamics of physical systems is often constrained to lower dimensional sub-spaces due to the presence of conserved quantities. Here we propose a method to learn and exploit such symmetry constraints building upon Hamiltonian Neural Networks. By enforcing cyclic coordinates with appropriate loss functions, we find that we can achieve improved accuracy on simple classical dynamics tasks. By fitting analytic formulae to the latent variables in our network we recover that our networks are utilizing conserved quantities such as (angular) momentum.
Cite
@article{arxiv.2104.14444,
title = {Improving Simulations with Symmetry Control Neural Networks},
author = {Marc Syvaeri and Sven Krippendorf},
journal= {arXiv preprint arXiv:2104.14444},
year = {2021}
}
Comments
7 pages, 2 figures, accepted as workshop paper at ICLR 2021 SimDL Workshop