A learning scheme to predict atomic forces and accelerate materials simulations

TL;DR

This paper introduces a machine learning approach to predict atomic forces, enabling faster materials simulations and extending quantum mechanical methods to larger scales.

Contribution

It presents a novel learned force field for aluminum that accelerates simulations and discusses pathways for systematic improvement and multi-element generalization.

Findings

Successfully predicts atomic forces in Al with high fidelity.

Enables simulations beyond current quantum mechanical length and time scales.

Framework for adaptive and multi-element force field development.

Abstract

The behavior of an atom in a molecule, liquid or solid is governed by the force it experiences. If the dependence of this vectorial force on the atomic chemical environment can be $learned$ efficiently with high-fidelity from benchmark reference results-using "big data" techniques, i.e., without resorting to actual functional forms-then this capability can be harnessed to enormously speed up $in \ silico$ materials simulations. The present contribution provides several examples of how such a $force$ field for Al can be used to go far beyond the length-scale and time-scale regimes accessible presently using quantum mechanical methods. It is argued that pathways are available to systematically and continuously improve the predictive capability of such a learned force field in an adaptive manner, and that this concept can be generalized to include multiple elements.