Synthesizability prediction of materials using deep learning

Abstract

Predicting the synthesizability of inorganic materials is one of the major challenges in accelerated material discovery. Considering simple thermodynamic decomposition stability due to its simplicity of computing is notorious for either producing too many candidates or missing important metastable materials. These results, however, are not unexcepted since the synthesizability is a complex phenomenon, and thermodynamic stability is just one contributor. Here, we suggest a machine-learned model to quantify the probability of synthesis based on partially supervised learning of materials database. We adapted the positive and unlabeled machine learning (PU-learning) by implementing graph convolutional neural network as a classifier, in which the model outputs synthesizability score. The model shows promising prediction accuracy for the test set of experimentally reported cases. The analysis shows that our model captures the chemical features for synthesizability beyond that is possible by thermodynamic stability. With the proposed data-driven metric of synthesizability score, high-throughput virtual screening and generative models can benefit significantly by effectively reducing the chemical space that needs to be explored experimentally in the future towards more rational materials design.