Machine learning assisted determination of electronic correlations from magnetic resonance

TL;DR

This paper demonstrates how machine learning techniques can analyze magnetic resonance data to extract detailed information about electronic correlations in strongly correlated materials, aiding experimental interpretation.

Contribution

It introduces machine learning models that classify and predict electronic correlation parameters from magnetic response data, advancing analysis methods for quantum materials.

Findings

Unsupervised learning classifies total interaction strength.

Supervised models predict spatial extent of correlations.

AI enhances interpretation of magnetic resonance experiments.

Abstract

In the presence of strong electronic spin correlations, the hyperfine interaction imparts long-range coupling between nuclear spins. Efficient protocols for the extraction of such complex information about electron correlations via magnetic response are not well known. Here, we study how machine learning can extract material parameters and help interpret magnetic response experiments. A low-dimensional representation that classifies the total interaction strength is discovered by unsupervised learning. Supervised learning generates models that predict the spatial extent of electronic correlations and the total interaction strength. Our work demonstrates the utility of artificial intelligence in the development of new probes of quantum systems, with applications to experimental studies of strongly correlated materials.