Adaptive coupling of a deep neural network potential to a classical force field

TL;DR

This paper introduces an adaptive modeling method that combines deep neural network potentials with classical force fields to improve accuracy and efficiency in molecular simulations, demonstrated on liquid water.

Contribution

The paper presents a novel adaptive coupling approach that seamlessly integrates deep neural network potentials with classical force fields in molecular simulations.

Findings

Effective decomposition of the system into regions with different modeling needs

Successful implementation of force interpolation and thermodynamics force in transition zones

Feasibility demonstrated on liquid water system

Abstract

An adaptive modeling method (AMM) that couples a deep neural network potential and a classical force field is introduced to address the accuracy-efficiency dilemma faced by the molecular simulation community. The AMM simulated system is decomposed into three types of regions. The first type captures the important phenomena in the system and requires high accuracy, for which we use the Deep Potential Molecular Dynamics (DeePMD) model in this work. The DeePMD model is trained to accurately reproduce the statistical properties of the ab initio molecular dynamics. The second type does not require high accuracy and a classical force field is used to describe it in an efficient way. The third type is used for a smooth transition between the first and the second types of regions. By using a force interpolation scheme and imposing a thermodynamics force in the transition region, we make the DeePMD region embedded in the AMM simulated system as if it were embedded in a system that is fully described by the accurate potential. A representative example of the liquid water system is used to show the feasibility and promise of this method.