Learning magnetization dynamics

Author(s)
Alexander Kovacs, Johann Fischbacher, Harald Oezelt, Markus Gusenbauer, Lukas Exl, Florian Bruckner, Dieter Suess, Thomas Schrefl
Abstract

Deep neural networks are used to model the magnetization dynamics in magnetic thin film elements. The magnetic states of a thin film element can be represented in a low dimensional space. With convolutional autoencoders a compression ratio of 1024:1 was achieved. Time integration can be performed in the latent space with a second network which was trained by solutions of the Landau-Lifshitz-Gilbert equation. Thus the magnetic response to an external field can be computed quickly.

Organisation(s)
Physics of Functional Materials
External organisation(s)
Donau-Universität Krems, Wolfgang Pauli Institute (WPI) Vienna
Journal
Journal of Magnetism and Magnetic Materials
Volume
491
No. of pages
6
ISSN
0304-8853
DOI
https://doi.org/10.1016/j.jmmm.2019.165548
Publication date
12-2019
Peer reviewed
Yes
Austrian Fields of Science 2012
Materials physics, Numerical mathematics, Machine learning
Keywords
ASJC Scopus subject areas
,
Portal url
https://ucris.univie.ac.at/portal/en/publications/learning-magnetization-dynamics(c7ad9c73-d0eb-4fd5-90a5-c38b5fdfac93).html