Predicting materials properties without crystal structure: Deep representation learning from stoichiometry
This work addresses a key bottleneck in materials discovery by enabling property predictions for materials with uncharacterized structures, which is incremental but impactful for accelerating materials science research.
The paper tackles the problem of predicting materials properties without requiring known crystal structures by developing a deep representation learning method that uses only stoichiometry as input, achieving lower errors with less data compared to state-of-the-art structure-agnostic methods.
Machine learning has the potential to accelerate materials discovery by accurately predicting materials properties at a low computational cost. However, the model inputs remain a key stumbling block. Current methods typically use descriptors constructed from knowledge of either the full crystal structure -- therefore only applicable to materials with already characterised structures -- or structure-agnostic fixed-length representations hand-engineered from the stoichiometry. We develop a machine learning approach that takes only the stoichiometry as input and automatically learns appropriate and systematically improvable descriptors from data. Our key insight is to treat the stoichiometric formula as a dense weighted graph between elements. Compared to the state of the art for structure-agnostic methods, our approach achieves lower errors with less data.