NEP-CG and NEP-AACG: Efficient coarse-grained and multiscale all-atom-coarse-grained neuroevolution potentials

TL;DR

This paper introduces a neuroevolution potential framework for creating accurate, transferable, and efficient coarse-grained and multiscale models, capable of handling noisy data and extrapolating beyond training conditions, demonstrated on liquids, materials, and nanostructures.

Contribution

The authors develop a novel method to generate low-noise training data and extend NEP models to multiscale all-atom and coarse-grained systems, improving accuracy and transferability.

Findings

NEP-CG accurately reproduces water densities across wide pressure ranges.

Distinguishing bead types reduces stress errors significantly.

NEP-AACG effectively models gold nanowire fracture at realistic strain rates.

Abstract

Machine-learned coarse-grained (CG) models often suffer from noisy training data, limiting their accuracy and transferability. We propose a method to generate low-noise training data based on the potential of mean force by constraining CG beads during atomistic simulations and accumulating time-averaged forces. Implemented within the neuroevolution potential (NEP) framework, our approach achieves training accuracy comparable to atomistic models trained on density functional theory data. For liquid water, the NEP-CG model accurately reproduces densities from 1 bar to 1 GPa, successfully extrapolating beyond the 0.5 GPa training limit, with a virial correction essential for the correct equation of state. For an anisotropic C$_{60}$ monolayer, distinguishing crystallographically distinct bead types reduces stress errors by an order of magnitude and captures directional thermal conductivity. We further introduce a multiscale NEP-AACG model integrating all-atom (AA) and CG degrees of freedom, demonstrated for gold nanowire fracture at an experimentally relevant strain rate. Computational speeds for NEP-CG models reach hundreds to thousands of ns/day using a single consumer-grade GPU. This work provides a robust framework for constructing accurate, transferable, and efficient CG models across diverse systems.