Machine Learning Diffusion Monte Carlo Forces

TL;DR

This paper demonstrates that machine learning models can accurately predict diffusion Monte Carlo forces, enabling high-precision molecular dynamics and geometry optimizations at a fraction of the usual computational cost.

Contribution

The study introduces a method to learn DMC forces from energy calculations using neural networks, overcoming challenges of noisy data and lack of explicit force data.

Findings

ML models reproduce experimental bond lengths and angles within a few percent.

DMC-based dynamics match high-accuracy PES predictions.

The approach reduces computational costs significantly.

Abstract

Diffusion Monte Carlo (DMC) is one of the most accurate techniques available for calculating the electronic properties of molecules and materials, yet it often remains a challenge to economically compute forces using this technique. As a result, ab initio molecular dynamics simulations and geometry optimizations that employ Diffusion Monte Carlo forces are often out of reach. One potential approach for accelerating the computation of "DMC forces" is to machine learn these forces from DMC energy calculations. In this work, we employ Behler-Parrinello Neural Networks to learn DMC forces from DMC energy calculations for geometry optimization and molecular dynamics simulations of small molecules. We illustrate the unique challenges that stem from learning forces without explicit force data and from noisy energy data by making rigorous comparisons of potential energy surface, dynamics, and optimization predictions among ab initio Density Functional Theory (DFT) simulations and machine learning models trained on DFT energies with forces, DFT energies without forces, and DMC energies without forces. We show for three small molecules - C2, H2O, and CH3Cl - that machine learned DMC dynamics can reproduce average bond lengths and angles within a few percent of known experimental results at a 100th of the typical cost. Our work describes a much-needed means of performing dynamics simulations on high-accuracy, DMC PESs and for generating DMC-quality molecular geometries given current algorithmic constraints.