Learning properties of ordered and disordered materials from multi-fidelity data

Predicting the properties of a material from the arrangement of its atoms is a fundamental goal in materials science. While machine learning has emerged in recent years as a new paradigm to provide rapid predictions of materials properties, their practical utility is limited by the scarcity of high-fidelity data. Here, we develop multi-fidelity graph networks as a universal approach to achieve accurate predictions of materials properties with small data sizes. As a proof of concept, we show that the inclusion of low-fidelity Perdew-Burke-Ernzerhof band gaps greatly enhances the resolution of latent structural features in materials graphs, leading to a 22-45% decrease in the mean absolute errors of experimental band gap predictions. We further demonstrate that learned elemental embeddings in materials graph networks provide a natural approach to model disorder in materials, addressing a fundamental gap in the computational prediction of materials properties.

Automated summary

Utilizing multi-fidelity graph networks as a universal approach can accurately predict materials properties with small data sizes. By incorporating low-fidelity Perdew-Burke-Ernzerhof band gaps, the resolution of latent structural features in materials graphs can be greatly enhanced. Learned elemental embeddings provide a natural approach to model disorder in materials.