Learning time-stepping by nonlinear dimensionality reduction to predict magnetization dynamics

DOI10.1016/J.CNSNS.2020.105205MaRDI QIDQ2204477zbMATH OpenOpenAlexWikidataFDO

Authors L. Exl, N. J. Mauser, T. Schrefl, Dieter Suess

Publication date 15 October 2020

Published in Communications in Nonlinear Science and Numerical Simulation (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1904.04215

zbMATH Keywords

kernel principal component analysiskernelizationmachine learningmicromagneticsnonlinear model order reduction

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Electromagnetic theory (general) (78A25) Finite difference methods applied to problems in optics and electromagnetic theory (78M20) Model reduction in optics and electromagnetic theory (78M34)

Abstract: We establish a time-stepping learning algorithm and apply it to predict the solution of the partial differential equation of motion in micromagnetism as a dynamical system depending on the external field as parameter. The data-driven approach is based on nonlinear model order reduction by use of kernel methods for unsupervised learning, yielding a predictor for the magnetization dynamics without any need for field evaluations after a data generation and training phase as precomputation. Magnetization states from simulated micromagnetic dynamics associated with different external fields are used as training data to learn a low-dimensional representation in so-called feature space and a map that predicts the time-evolution in reduced space. Remarkably, only two degrees of freedom in feature space were enough to describe the nonlinear dynamics of a thin-film element. The approach has no restrictions on the spatial discretization and might be useful for fast determination of the response to an external field.

Recommendations

• Data-driven learning of nonautonomous systems
• Kernel-based prediction of non-Markovian time series
• Reduced operator inference for nonlinear partial differential equations
• Machine learning for fast and reliable solution of time-dependent differential equations
• A Gauss-Seidel projection method for micromagnetics simulations

Cites work

• Geometrical integration of Landau-Lifshitz-Gilbert equation based on the mid-point rule
• scientific article; zbMATH DE number 193745 (Why is no real title available?)
• scientific article; zbMATH DE number 3210973 (Why is no real title available?)
• Kernel methods in machine learning
• Non-uniform FFT for the finite element computation of the micromagnetic scalar potential
• Scikit-learn: machine learning in Python
• The extrapolated explicit midpoint scheme for variable order and step size controlled integration of the Landau-Lifschitz-Gilbert equation

Cited in

(2)

• Prediction of magnetization dynamics in a reduced dimensional feature space setting utilizing a low-rank kernel method
• Constraint free physics-informed machine learning for micromagnetic energy minimization

Describes a project that uses

Uses Software

• OOMMF
• Scikit

This page was built for publication: Learning time-stepping by nonlinear dimensionality reduction to predict magnetization dynamics

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2204477)