Beyond potential energy surface benchmarking: a complete application of machine learning to chemical reactivity

TL;DR

This paper demonstrates an active learning approach using metadynamics to train machine learning models for predicting energies and forces in hydrogen combustion, improving PES completeness and enabling efficient, real-world reactive chemistry simulations.

Contribution

It introduces a negative design data acquisition strategy and an active learning workflow based on metadynamics to create more complete ML potential energy surfaces for reactive systems.

Findings

ML PES models require negative design data for completeness

Active learning accelerates PES training and reduces computational cost

Hybrid ML-physics models accurately predict reaction mechanisms at finite conditions

Abstract

We train an equivariant machine learning model to predict energies and forces for a real-world study of hydrogen combustion under conditions of finite temperature and pressure. This challenging case for reactive chemistry illustrates that ML learned potential energy surfaces (PESs) are always incomplete as they are overly reliant on chemical intuition of what data is important for training, i.e. stable or metastable energy states. Instead we show here that a negative design data acquisition strategy is necessary to create a more complete ML model of the PES, since it must also learn avoidance of unforeseen high energy intermediates or even unphysical energy configurations. Because this type of data is unintuitive to create, we introduce an active learning workflow based on metadynamics that samples a lower dimensional manifold within collective variables that efficiently creates highly variable energy configurations for further ML training. This strategy more rapidly completes the ML PES such that deviations among query by committee ML models helps to now signal occasional calls to the external ab initio data source to further molecular dynamics in time without need for retraining the ML model. With the hybrid ML-physics model we predict the change in transition state and/or reaction mechanism at finite temperature and pressure for hydrogen combustion, thereby delivering on the promise of real application work using ML trained models of an ab initio PES with two orders of magnitude reduction in cost.