CTGNN: Crystal Transformer Graph Neural Network for Crystal Material Property Prediction

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Du, Zijian, Jin, Luozhijie, Shu, Le, Cen, Yan, Xu, Yuanfeng, Mei, Yongfeng, Zhang, Hao
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914802449776640
author Du, Zijian
Jin, Luozhijie
Shu, Le
Cen, Yan
Xu, Yuanfeng
Mei, Yongfeng
Zhang, Hao
author_facet Du, Zijian
Jin, Luozhijie
Shu, Le
Cen, Yan
Xu, Yuanfeng
Mei, Yongfeng
Zhang, Hao
contents The combination of deep learning algorithm and materials science has made significant progress in predicting novel materials and understanding various behaviours of materials. Here, we introduced a new model called as the Crystal Transformer Graph Neural Network (CTGNN), which combines the advantages of Transformer model and graph neural networks to address the complexity of structure-properties relation of material data. Compared to the state-of-the-art models, CTGNN incorporates the graph network structure for capturing local atomic interactions and the dual-Transformer structures to model intra-crystal and inter-atomic relationships comprehensively. The benchmark carried on by the proposed CTGNN indicates that CTGNN significantly outperforms existing models like CGCNN and MEGNET in the prediction of formation energy and bandgap properties. Our work highlights the potential of CTGNN to enhance the performance of properties prediction and accelerates the discovery of new materials, particularly for perovskite materials.
format Preprint
id arxiv_https___arxiv_org_abs_2405_11502
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle CTGNN: Crystal Transformer Graph Neural Network for Crystal Material Property Prediction
Du, Zijian
Jin, Luozhijie
Shu, Le
Cen, Yan
Xu, Yuanfeng
Mei, Yongfeng
Zhang, Hao
Materials Science
Computational Physics
The combination of deep learning algorithm and materials science has made significant progress in predicting novel materials and understanding various behaviours of materials. Here, we introduced a new model called as the Crystal Transformer Graph Neural Network (CTGNN), which combines the advantages of Transformer model and graph neural networks to address the complexity of structure-properties relation of material data. Compared to the state-of-the-art models, CTGNN incorporates the graph network structure for capturing local atomic interactions and the dual-Transformer structures to model intra-crystal and inter-atomic relationships comprehensively. The benchmark carried on by the proposed CTGNN indicates that CTGNN significantly outperforms existing models like CGCNN and MEGNET in the prediction of formation energy and bandgap properties. Our work highlights the potential of CTGNN to enhance the performance of properties prediction and accelerates the discovery of new materials, particularly for perovskite materials.
title CTGNN: Crystal Transformer Graph Neural Network for Crystal Material Property Prediction
topic Materials Science
Computational Physics
url https://arxiv.org/abs/2405.11502