PREDICTING OF ELASTIC AND THERMAL PROPERTIES OF METALLIC GLASSES USING GAUSSIAN PROCESS REGRESSION

Abstract

In this study, we used Gaussian process regression (GPR) to predict the elastic moduli and glass transition temperature of metallic glasses based on their atomic composition. The results show that GPR is an effective method for predicting these properties, with high accuracy and low error compared to experimental data. We also analyze the contributions of individual elements to the properties of metallic glasses, providing insight into their underlying physical mechanisms. This study demonstrates the potential of GPR for predicting the properties of complex materials, and highlights the importance of understanding the relationships between composition and properties in metallic glasses