Identifying magnetic antiskyrmions while they form with convolutional neural networks

Chiral magnets have attracted a large amount of research interest in recent years because they support variety topological defects, such as skyrmions and bimerons, allow for their observation manipulation through several techniques. They also wide range applications the field spintronics, particularly developing new technologies memory storage devices. However, vast data generated these experimental theoretical studies requires adequate tools, among which machine learning is crucial. We use Convolutional Neural Network (CNN) to identify relevant features thermodynamical phases chiral magnets, including (anti-)skyrmions, helical ferromagnetic states. flexible multi-label classification framework that can correctly classify states different are mixed. then train CNN predict final state from snapshots intermediate lattice Monte Carlo simulation. The trained model allows identifying reliably early formation process. Thus, significantly speed up large-scale simulations 3D materials been bottleneck quantitative so far. Moreover, this approach be applied identification mixed emerging real-world images magnets.