TY  - CONF
AU  - Aldarawsheh, Amal
AU  - Alia, Ahmed
AU  - Blügel, Stefan
TI  - CNN-based Classification of Magnetic States from Atomistic Simulations in thin films
M1  - FZJ-2025-03185
PY  - 2025
AB  - The identification and classification of different magnetic states are essential for understanding the complex behavior of magnetic systems. Traditional approaches that rely on handcrafted features or manual inspection often fall short, particularly when dealing with subtle or topologically complex spin textures. In this study, we present a fully automated deep learning model that employs an EfficientNetV1B0 Convolutional Neural Network (CNN) to classify nine distinct magnetic states, including both FM and, for the first time, AFM spin textures such as AFM skyrmions and AFM stripe domains. The spin configurations are generated through atomistic spin dynamics simulations using the \textit{Spirit} code, then visualized with VFRendering script. Our model achieves a classification accuracy and F1-score of 99\%, significantly outperforming established CNN baselines and demonstrating exceptional capability in distinguishing closely related magnetic states.
T2  - Virtual Materials Design 2025 - Karlsruhe
CY  - 2 Jun 2025 - 5 Jun 2025, Karlsuhe (Germany)
Y2  - 2 Jun 2025 - 5 Jun 2025
M2  - Karlsuhe, Germany
LB  - PUB:(DE-HGF)6
UR  - https://juser.fz-juelich.de/record/1044424
ER  -