Machine learning for electronically excited states of molecules

TL;DR

This review explores how machine learning accelerates and enhances the study of electronically excited states in molecules, impacting fields like photochemistry, photophysics, and material science.

Contribution

It provides a comprehensive overview of machine learning applications in excited-state molecular simulations, highlighting recent advances and challenges.

Findings

Machine learning speeds up excited-state calculations.

ML improves accuracy in absorption spectra predictions.

Challenges include data quality and method transferability.

Abstract

Electronically excited states of molecules are at the heart of photochemistry, photophysics, as well as photobiology and also play a role in material science. Their theoretical description requires highly accurate quantum chemical calculations, which are computationally expensive. In this review, we focus on how machine learning is employed not only to speed up such excited-state simulations but also how this branch of artificial intelligence can be used to advance this exciting research field in all its aspects. Discussed applications of machine learning for excited states include excited-state dynamics simulations, static calculations of absorption spectra, as well as many others. In order to put these studies into context, we discuss the promises and pitfalls of the involved machine learning techniques. Since the latter are mostly based on quantum chemistry calculations, we also provide a short introduction into excited-state electronic structure methods, approaches for nonadiabatic dynamics simulations and describe tricks and problems when using them in machine learning for excited states of molecules.