|About this Abstract
||Materials Science & Technology 2019
||Ceramics and Glasses Simulations and Machine Learning
||Machine Learning to Predict the Elastic Properties of Glasses
||N. M. Anoop Krishnan, Nishank Goyal, Divyarth Prakash Saxena, Sourabh Kumar Singh, Suresh Bishnoi, Ravinder Ravinder, Hariprasad Kodamana
|On-Site Speaker (Planned)
||N. M. Anoop Krishnan
Glasses are archetypical disordered materials that are used ubiquiitously in daily life applications. Due to their unique disordered structure, predicting the composition–property relationships in glasses are extremely challenging. Herein, using advanced machine learning (ML) methods, we predict the elastic properties of silicate glasses. In particular, we use deep neural networks (NN) and Gaussian process regression (GPR) to model the elastic properties such as Young's modulus, Poisson's ratio, and density by training against a large dataset obtained from previous experiments. We show that the ML models can predict elastic properties with high degree of accuracy and reliability. Development of such models is imperative to accelerate the design of novel functional glasses for practical applications.