Cargando…

Neural network potentials for chemistry: concepts, applications and prospects

Artificial Neural Networks (NN) are already heavily involved in methods and applications for frequent tasks in the field of computational chemistry such as representation of potential energy surfaces (PES) and spectroscopic predictions. This perspective provides an overview of the foundations of neu...

Descripción completa

Detalles Bibliográficos
Autores principales: Käser, Silvan, Vazquez-Salazar, Luis Itza, Meuwly, Markus, Töpfer, Kai
Formato: Online Artículo Texto
Lenguaje:English
Publicado: RSC 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9923808/
https://www.ncbi.nlm.nih.gov/pubmed/36798879
http://dx.doi.org/10.1039/d2dd00102k
_version_ 1784887785884221440
author Käser, Silvan
Vazquez-Salazar, Luis Itza
Meuwly, Markus
Töpfer, Kai
author_facet Käser, Silvan
Vazquez-Salazar, Luis Itza
Meuwly, Markus
Töpfer, Kai
author_sort Käser, Silvan
collection PubMed
description Artificial Neural Networks (NN) are already heavily involved in methods and applications for frequent tasks in the field of computational chemistry such as representation of potential energy surfaces (PES) and spectroscopic predictions. This perspective provides an overview of the foundations of neural network-based full-dimensional potential energy surfaces, their architectures, underlying concepts, their representation and applications to chemical systems. Methods for data generation and training procedures for PES construction are discussed and means for error assessment and refinement through transfer learning are presented. A selection of recent results illustrates the latest improvements regarding accuracy of PES representations and system size limitations in dynamics simulations, but also NN application enabling direct prediction of physical results without dynamics simulations. The aim is to provide an overview for the current state-of-the-art NN approaches in computational chemistry and also to point out the current challenges in enhancing reliability and applicability of NN methods on a larger scale.
format Online
Article
Text
id pubmed-9923808
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher RSC
record_format MEDLINE/PubMed
spelling pubmed-99238082023-02-14 Neural network potentials for chemistry: concepts, applications and prospects Käser, Silvan Vazquez-Salazar, Luis Itza Meuwly, Markus Töpfer, Kai Digit Discov Chemistry Artificial Neural Networks (NN) are already heavily involved in methods and applications for frequent tasks in the field of computational chemistry such as representation of potential energy surfaces (PES) and spectroscopic predictions. This perspective provides an overview of the foundations of neural network-based full-dimensional potential energy surfaces, their architectures, underlying concepts, their representation and applications to chemical systems. Methods for data generation and training procedures for PES construction are discussed and means for error assessment and refinement through transfer learning are presented. A selection of recent results illustrates the latest improvements regarding accuracy of PES representations and system size limitations in dynamics simulations, but also NN application enabling direct prediction of physical results without dynamics simulations. The aim is to provide an overview for the current state-of-the-art NN approaches in computational chemistry and also to point out the current challenges in enhancing reliability and applicability of NN methods on a larger scale. RSC 2022-12-21 /pmc/articles/PMC9923808/ /pubmed/36798879 http://dx.doi.org/10.1039/d2dd00102k Text en This journal is © The Royal Society of Chemistry https://creativecommons.org/licenses/by/3.0/
spellingShingle Chemistry
Käser, Silvan
Vazquez-Salazar, Luis Itza
Meuwly, Markus
Töpfer, Kai
Neural network potentials for chemistry: concepts, applications and prospects
title Neural network potentials for chemistry: concepts, applications and prospects
title_full Neural network potentials for chemistry: concepts, applications and prospects
title_fullStr Neural network potentials for chemistry: concepts, applications and prospects
title_full_unstemmed Neural network potentials for chemistry: concepts, applications and prospects
title_short Neural network potentials for chemistry: concepts, applications and prospects
title_sort neural network potentials for chemistry: concepts, applications and prospects
topic Chemistry
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9923808/
https://www.ncbi.nlm.nih.gov/pubmed/36798879
http://dx.doi.org/10.1039/d2dd00102k
work_keys_str_mv AT kasersilvan neuralnetworkpotentialsforchemistryconceptsapplicationsandprospects
AT vazquezsalazarluisitza neuralnetworkpotentialsforchemistryconceptsapplicationsandprospects
AT meuwlymarkus neuralnetworkpotentialsforchemistryconceptsapplicationsandprospects
AT topferkai neuralnetworkpotentialsforchemistryconceptsapplicationsandprospects