Machine Learning for Atomic Forces in a Crystalline Solid: Transferability to Various Temperatures

TL;DR

This paper demonstrates that a machine learning model trained on high-temperature quantum data can accurately predict atomic forces in crystalline silicon across a wide temperature range, showing transferability to larger systems.

Contribution

The study introduces a transferability approach for ML models trained at high temperatures to predict atomic forces across various temperatures and system sizes.

Findings

Force prediction errors below 2% across 300-1650 K

Model transferability to larger systems verified

Effective use of kernel ridge regression with atomic fingerprints

Abstract

Recently, machine learning has emerged as an alternative, powerful approach for predicting quantum-mechanical properties of molecules and solids. Here, using kernel ridge regression and atomic fingerprints representing local environments of atoms, we trained a machine-learning model on a crystalline silicon system in order to directly predict the atomic forces at a wide range of temperatures. Our idea is to construct a machine-learning model using a quantum-mechanical data set taken from canonical-ensemble simulations at a higher temperature, or an upper bound of the temperature range. With our model, the force prediction errors were about 2% or smaller with respect to the corresponding force ranges, in the temperature region between 300 and 1650 K. We also verified the applicability to a larger system, ensuring the transferability with respect to system size.