Article (Périodiques scientifiques)
Extending atomic decomposition and many-body representation with a chemistry-motivated approach to machine learning potentials.
Yu, Qi; Ma, Ruitao; Qu, Chen et al.
2025In Nature computational science
Peer reviewed
 

Documents


Texte intégral
Nat_Comput_Sci_2025.pdf
Postprint Auteur (1.63 MB)
Télécharger

Tous les documents dans ORBilu sont protégés par une licence d'utilisation.

Envoyer vers



Détails



Mots-clés :
Atomic decomposition; Body representations; Condensed phase; Energy; Forcefields; Invariant polynomials; Learning potential; Machine-learning; Many body; Permutationally invariant; Computer Science (miscellaneous); Computer Science Applications; Computer Networks and Communications
Résumé :
[en] Most widely used machine learning potentials for condensed-phase applications rely on many-body permutationally invariant polynomial or atom-centered neural networks. However, these approaches face challenges in achieving chemical interpretability in atomistic energy decomposition and fully matching the computational efficiency of traditional force fields. Here we present a method that combines aspects of both approaches and balances accuracy and force-field-level speed. This method utilizes a monomer-centered representation, where the potential energy is decomposed into the sum of chemically meaningful monomeric energies. The structural descriptors of monomers are described by one-body and two-body effective interactions, enforced by appropriate sets of permutationally invariant polynomials as inputs to the feed-forward neural networks. Systematic assessments of models for gas-phase water trimer, liquid water, methane-water cluster and liquid carbon dioxide are performed. The improved accuracy, efficiency and flexibility of this method have promise for constructing accurate machine learning potentials and enabling large-scale quantum and classical simulations for complex molecular systems.
Disciplines :
Chimie
Auteur, co-auteur :
Yu, Qi ;  Department of Chemistry, Fudan University, Shanghai, China. qi_yu@fudan.edu.cn ; Shanghai Innovation Institute, Shanghai, China. qi_yu@fudan.edu.cn
Ma, Ruitao;  Department of Chemistry, Fudan University, Shanghai, China
Qu, Chen;  Independent Researcher, Toronto, Ontario, Canada
Conte, Riccardo ;  Dipartimento di Chimica, Università degli Studi di Milano, Milan, Italy
NANDI, Apurba  ;  University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Physics and Materials Science (DPHYMS)
Pandey, Priyanka;  Department of Chemistry and Cherry L. Emerson Center for Scientific Computation, Emory University, Atlanta, GA, USA
Houston, Paul L;  Department of Chemistry and Chemical Biology, Cornell University, Ithaca, NY, USA
Zhang, Dong H ;  State Key Laboratory of Molecular Reaction Dynamics, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian, China
Bowman, Joel M ;  Department of Chemistry and Cherry L. Emerson Center for Scientific Computation, Emory University, Atlanta, GA, USA
Co-auteurs externes :
yes
Langue du document :
Anglais
Titre :
Extending atomic decomposition and many-body representation with a chemistry-motivated approach to machine learning potentials.
Date de publication/diffusion :
14 avril 2025
Titre du périodique :
Nature computational science
ISSN :
2662-8457
eISSN :
2662-8457
Maison d'édition :
Springer Nature, Etats-Unis
Peer reviewed :
Peer reviewed
Subventionnement (détails) :
Q.Y. and D.H.Z. acknowledge the support from National Natural Science Foundation of China (grant numbers 22473030 and 22288201). J.M.B. acknowledges support from NASA grant (80NSSC22K1167). R.C. thanks Universit\u00E0 degli Studi di Milano for financial support under grant PSR2022_DIP_005_PI_RCONT.
Disponible sur ORBilu :
depuis le 22 avril 2025

Statistiques


Nombre de vues
77 (dont 1 Unilu)
Nombre de téléchargements
79 (dont 0 Unilu)

citations Scopus®
 
5
citations Scopus®
sans auto-citations
3
OpenCitations
 
0
citations OpenAlex
 
4

Bibliographie


Publications similaires



Contacter ORBilu