University of Vienna team uses physics-guided AI to map ultrathin magnet patterns
Researchers at the University of Vienna have developed a physics-guided artificial intelligence method to reconstruct internal magnetic structures in ultrathin materials based on external field measurements.
Solving the inverse problem for magnets
Scientists previously struggled because different internal patterns could create identical external magnetic fields. The new approach builds physical laws directly into machine learning algorithms to resolve this ambiguity. This technique allows researchers to see realistic magnetic structures that standard microscopes cannot detect.
Determining sensor distance automatically
The software also calculates the precise distance between the sensor and the sample without extra input. Such materials only a few atomic layers thick are important for future data storage devices. The team collaborated with colleagues in Dresden, Singapore, and Ulsan to create this solution.
Open source software release
The study appears in Physical Review Letters and the code is available as open source. This tool helps build faster and more energy-efficient computers by understanding nanoscale magnetic materials better.
Reported by one outlet
Only one outlet has published this. Nothing here has been checked against a second report, so read it as that outlet's account and follow the link below for the original.
Reported by
1 independent outlet. Headline as published. Links open the original report.