Graph neural network-based structural classification of glass-forming liquids and its interpretation via self-attention mechanism
The Journal of Chemical Physics · 14.07.2025 · 163 (2)
DOI: 10.1063/5.0277279 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Journal of Chemical Physics, online erschienen am 10.07.2025. Verfasst von Kohei Yoshikawa und 4 weiteren Autor:innen. Beteiligte Einrichtung: Division of Chemical Engineering, Department of Materials Engineering Science, Graduate School of Engineering Science, The University of Osaka, Toyon…. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 74 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Japan Society for the Promotion of Science, Ministry of Education, Culture, Sports, Science and Technology.
Abstract (Originalsprache)
Glass-forming liquids exhibit slow dynamics below their melting temperatures, maintaining an amorphous structure reminiscent of normal liquids. Distinguishing microscopic structures in the supercooled and high-temperature regimes remains a debated topic. Building on recent advances in machine learning, particularly Graph Neural Networks (GNNs), our study automatically extracts features, unveiling fundamental mechanisms driving structural changes at varying temperatures. We employ the self-attention mechanism to generate attention coefficients that quantify the importance of connections between graph nodes, providing insights into the rationale behind GNN predictions. By exploring structural changes with decreasing temperature through the GNN + self-attention using physically defined structural descriptors, including the bond-orientational order parameter, Voronoi cell volume, and coordination number, we identify strong correlations between high attention coefficients and more disordered structures as a key indicator of variations in glass-forming liquids.
Autor:innen und Institutionen
- Division of Chemical Engineering, Department of Materials Engineering Science, Graduate School of Engineering Science, The University of Osaka, Toyonaka, Osaka 560-8531, Japan.
ORCID-Profile: Kohei Yoshikawa, Kang Kim, Nobuyuki Matubayasi
Förderung
- Japan Society for the Promotion of Science
- Ministry of Education, Culture, Sports, Science and Technology
Zitieren
APA
Yoshikawa, K., Yano, K., Goto, S., Kim, K., & Matubayasi, N. (2025). Graph neural network-based structural classification of glass-forming liquids and its interpretation via self-attention mechanism. The Journal of Chemical Physics, 163(2). https://doi.org/10.1063/5.0277279
Vancouver
Yoshikawa K, Yano K, Goto S, Kim K, Matubayasi N. Graph neural network-based structural classification of glass-forming liquids and its interpretation via self-attention mechanism. The Journal of Chemical Physics. 2025;163(2). doi:10.1063/5.0277279
BibTeX
@article{Yoshikawa2025,
author = {Yoshikawa, Kohei and Yano, Kentaro and Goto, Shota and Kim, Kang and Matubayasi, Nobuyuki},
title = {Graph neural network-based structural classification of glass-forming liquids and its interpretation via self-attention mechanism},
journal = {The Journal of Chemical Physics},
year = {2025},
volume = {163},
number = {2},
doi = {10.1063/5.0277279}
}
EUVIMED