Glass‐Forming‐Region‐Informed Machine Learning Reveals Quantitative Criteria for Network Glass Formation
Journal of the American Ceramic Society · 2026-07 · 109 (7)
DOI: 10.1111/jace.70980 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of the American Ceramic Society, online erschienen am 07.07.2026. Verfasst von Yao Ji und 2 weiteren Autor:innen. Die Arbeit nennt 47 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
The accurate prediction of glass formation in inorganic materials remains a formidable challenge in condensed matter science. Current machine learning approaches often rely on glass‐forming ability datasets biased toward successful glass formers, compromising their generalizability and interpretability. Here, we present a comprehensive glass‑forming‐region‐informed machine learning (GFRI‐ML) framework, integrating both composition and physical‐property descriptors to quantitatively predict glass formation across diverse oxide systems. Our curated dataset comprises 44 753 oxide glass samples from 2771 ternary systems, specifically incorporating negative (non‐glass‐forming) instances to eliminate selection bias. We developed two hierarchical predictive models: (i) a Composition‐GFR model using only compositional features, achieving a cross‑validation accuracy of 90.6%, and (ii) a Property‑GFR model incorporating 51 weighted physical‐chemical descriptors, reaching 94.6% accuracy. Both models demonstrated superior generalization performance on independent, held‐out experimental systems. Furthermore, Shapley Additive Explanations analyses were employed to elucidate generalizable physical laws governing vitrification, revealing critical thresholds within the kinetic, thermodynamic, and topological dimensions. This work provides an interpretable, high‐fidelity tool for glass prediction and establishes a quantitative, physics‐grounded foundation for the accelerated discovery of novel inorganic glasses.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Ji, Y., Dong, S., & Zhang, Q. (2026). Glass‐Forming‐Region‐Informed Machine Learning Reveals Quantitative Criteria for Network Glass Formation. Journal of the American Ceramic Society, 109(7). https://doi.org/10.1111/jace.70980
Vancouver
Ji Y, Dong S, Zhang Q. Glass‐Forming‐Region‐Informed Machine Learning Reveals Quantitative Criteria for Network Glass Formation. Journal of the American Ceramic Society. 2026;109(7). doi:10.1111/jace.70980
BibTeX
@article{Ji2026,
author = {Ji, Yao and Dong, Shuangli and Zhang, Qinyuan},
title = {Glass‐Forming‐Region‐Informed Machine Learning Reveals Quantitative Criteria for Network Glass Formation},
journal = {Journal of the American Ceramic Society},
year = {2026},
volume = {109},
number = {7},
doi = {10.1111/jace.70980}
}
EUVIMED