Predicting oxide glass properties with low complexity neural network and physical and chemical descriptors

Abstract

Due to disordered structure, glasses present a unique challenge in predicting the composition–property relationships. Here, we present a low-complexity neural network (LCNN) that provides improved performance in predicting the properties of oxide glasses. In addition, we combine the LCNN with physical and chemical descriptors that allow the development of universal models that can provide predictions for components beyond the training set. By training on a large dataset (∼50,000) of glass components, we show the LCNN outperforms state-of-the-art algorithms such as XGBoost. In addition, we interpret the LCNN models using Shapely additive explanations to gain insights into the role played by the descriptors in governing the property. Finally, we demonstrate the universality of the LCNN models by predicting the properties of glasses with new components that were not present in the original training set. Altogether, the present approach provides a promising direction towards the accelerated discovery of novel glass compositions.