Deep learning statistical defect models on magnetic material dynamic and static properties

TL;DR

This paper introduces a deep learning framework that models magnetic material defects, integrating physics-informed neural networks to predict physical properties and defect thresholds, aiding material discovery.

Contribution

It develops a novel combination of deep learning and physics-informed models to predict magnetic properties considering defects, advancing material modeling techniques.

Findings

Deep learning models accurately predict dispersion relations with defect parameters.

Physics-informed neural networks incorporate physical constraints into predictions.

The approach facilitates discovery of new magnetic materials and defect thresholds.

Abstract

The modeling of realistic magnetic materials requires the inclusion of defects. Based on the pseudospectral Landau-Lifshitz description of magnetisation dynamics, we propose a statistical model that takes into account defects, specifically vacancies. This statistical model can be integrated with deep learning techniques that correlate defect thresholds with relevant physical observables. We develop a convolutional neural network and a physics-informed neural network combined with theory of functional connections to predict the dispersion relation given defect parameters and physical constraints. A two-branch convolutional neural network is developed to predict domain-wall widths depending on defects threshold, taking into account the spatial profile and domain-wall width separately to achieve a prediction. The proposed physics-informed approaches leverage deep-learning and achieve statistical predictions measured in physical units. This is a stepping stone towards the discovery of new materials and the determination of minimal defect thresholds required for desired dynamics, states, or topological textures.