Predicting Young's Modulus of Glasses with Sparse Datasets using Machine Learning

TL;DR

This paper demonstrates that Gaussian process regression (GPR) effectively predicts Young's modulus of silicate glasses from sparse datasets, outperforming neural networks and providing reliable uncertainty estimates, thus aiding material design.

Contribution

The study introduces GPR as a superior alternative to neural networks for predicting material properties from limited data, with built-in uncertainty quantification.

Findings

GPR outperforms neural networks on sparse datasets.

GPR provides quantitative uncertainty bounds.

GPR avoids overfitting in small data regimes.

Abstract

Machine learning (ML) methods are becoming popular tools for the prediction and design of novel materials. In particular, neural network (NN) is a promising ML method, which can be used to identify hidden trends in the data. However, these methods rely on large datasets and often exhibit overfitting when used with sparse dataset. Further, assessing the uncertainty in predictions for a new dataset or an extrapolation of the present dataset is challenging. Herein, using Gaussian process regression (GPR), we predict Young's modulus for silicate glasses having sparse dataset. We show that GPR significantly outperforms NN for sparse dataset, while ensuring no overfitting. Further, thanks to the nonparametric nature, GPR provides quantitative bounds for the reliability of predictions while extrapolating. Overall, GPR presents an advanced ML methodology for accelerating the development of novel functional materials such as glasses.