Ultra-fast interpretable machine-learning potentials

TL;DR

This paper introduces ultra-fast, interpretable machine-learning potentials that combine physical insights with computational efficiency, enabling accurate large-scale atomistic simulations at speeds comparable to traditional empirical models.

Contribution

The authors develop a novel machine-learning potential framework using B-spline basis and regularized linear regression, achieving high accuracy and interpretability while being significantly faster than existing ML potentials.

Findings

Potential accurately retrieves empirical data

Predicted properties closely match DFT results

Achieves 2-4 orders of magnitude speedup

Abstract

All-atom dynamics simulations are an indispensable quantitative tool in physics, chemistry, and materials science, but large systems and long simulation times remain challenging due to the trade-off between computational efficiency and predictive accuracy. To address this challenge, we combine effective two- and three-body potentials in a cubic B-spline basis with regularized linear regression to obtain machine-learning potentials that are physically interpretable, sufficiently accurate for applications, as fast as the fastest traditional empirical potentials, and two to four orders of magnitude faster than state-of-the-art machine-learning potentials. For data from empirical potentials, we demonstrate exact retrieval of the potential. For data from density functional theory, the predicted energies, forces, and derived properties, including phonon spectra, elastic constants, and melting points, closely match those of the reference method. The introduced potentials might contribute towards accurate all-atom dynamics simulations of large atomistic systems over long time scales.