Crystal Structure Based AI/ML Approach for Material Magnetic Ordering Prediction

AI/ML models are useful for the prediction of materials' properties. Recently, there has been great progress in developing Neural Network models to predict the properties of solids, molecules, and crystals, avoiding expensive and time-consuming laboratory testing. This paper involves both Graph Neural Network (GNN) and Large Language Model (LLM)-based models to predict material magnetic ordering in four classes: Non-magnetic, Ferromagnetic, Antiferromagnetic, and Ferrimagnetic. Material crystal structure information is fed into a GNN and also converted to semantic text descriptions and tokenized as input to a LLM. The models are trained, validated, and tested on a dataset which is a combination of Materials Project dataset, Magndata dataset, and a set of uranium compounds with magnetic ordering determined in laboratory testing, with an 80/10/10 (training/validation/testing) data split. The study features two stage training process to train feature extraction and classification separately, which is necessary to handle class imbalance in the dataset due to limited availability of data for certain classes (AFM and FiM). In the first stage, the feature extraction is trained; in the second stage, the feature extraction is frozen and the classification head is trained for the final classification task. Based on the two stage training, the proposed LLM and ALIGNN-LLM multi-modal approaches can achieve an average 90% overall accuracy on the testing dataset across the four classes, and high (80% - 90%) per-class precision, recall and F1-macro scores.