How graph neural network interatomic potentials extrapolate: Role of the message-passing algorithm

Graph neural network interatomic potentials (GNN-IPs) are gaining significant attention due to their capability of learning from large datasets. Specifically, universal interatomic potentials based on GNN, usually trained with crystalline geometries, often exhibit remarkable extrapolative behavior t...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	The Journal of chemical physics 2024-12, Vol.161 (24)
1. Verfasser:	Kang, Sungwoo
Format:	Artikel
Sprache:	eng
Schlagworte:	Algorithms Density functional theory Extrapolation Graph neural networks IP (Internet Protocol) Machine learning Message passing Neural networks
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!