Atomistic Simulations for Reactions and Spectroscopy in the Era of Machine Learning -- Quo Vadis?

TL;DR

This paper reviews the current state and future prospects of atomistic simulations combined with machine learning techniques, highlighting their potential to enhance molecular dynamics understanding in various phases.

Contribution

It provides a comprehensive overview of how machine learning is integrated into atomistic simulations and discusses open questions and future directions in the field.

Findings

Machine learning enhances the accuracy of energy functions in simulations.

Integration of ML with atomistic simulations improves molecular dynamics insights.

Open questions include scalability and transferability of ML models in simulations.

Abstract

Atomistic simulations using accurate energy functions can provide molecular-level insight into functional motions of molecules in the gas- and in the condensed phase. Together with recently developed and currently pursued efforts in integrating and combining this with machine learning techniques provides a unique opportunity to bring such dynamics simulations closer to reality. This perspective delineates the present status of the field from efforts of others in the field and some of your own work and discusses open questions and future prospects.