3D-structure-attention graph neural network for crystals and materials. (3rd June 2022)
- Record Type:
- Journal Article
- Title:
- 3D-structure-attention graph neural network for crystals and materials. (3rd June 2022)
- Main Title:
- 3D-structure-attention graph neural network for crystals and materials
- Authors:
- Lin, Xuanjie
Jiang, Hantong
Wang, Liquan
Ren, Yongsheng
Ma, Wenhui
Zhan, Shu - Abstract:
- Abstract : Machine learning has been widely used in physics and chemistry. As a deep learning method based on graph domain analysis, graph neural networks (GNNs) have natural advantages in predicting material properties. We find that most existing models focus on the topological relationship between atoms without considering the specific positions. However, 3D-spatial distribution is the key to affecting the atomic state and interaction relationship, which has a decisive impact on the material properties. Here, we present a 3D-structure-attention graph neural network (3SAGNN) model, introducing the attention mechanism. The model focuses on the critical areas in the material 3D structure that significantly impact the prediction properties to effectively improve the accuracy of material properties prediction. We prove that the performance of 3SAGNN on a variety of datasets outperforms prior ML models, such as CGCNN. Our proposed model was tested on 36, 000 inorganic materials dataset, 20, 000 Pt nanocluster dataset, 18, 000 porous materials, and 37, 000 alloy surface reactions. The experimental results show that 3SAGNN can predict formation energies, total energies, band gaps, and surface catalytic properties more accurately and quickly than density functional theory. Finally, we improve the interpretability of the model through visualisation and show the working mechanism of the network. GRAPHICAL ABSTRACT: UF0001
- Is Part Of:
- Molecular physics. Volume 120:Number 11(2022)
- Journal:
- Molecular physics
- Issue:
- Volume 120:Number 11(2022)
- Issue Display:
- Volume 120, Issue 11 (2022)
- Year:
- 2022
- Volume:
- 120
- Issue:
- 11
- Issue Sort Value:
- 2022-0120-0011-0000
- Page Start:
- Page End:
- Publication Date:
- 2022-06-03
- Subjects:
- Graph neural network -- deep learning for materials science -- machine learning -- molecular property prediction
Molecules -- Periodicals
Chemistry, Physical and theoretical -- Periodicals
Molécules -- Périodiques
Chimie physique et théorique -- Périodiques
539.6.05 - Journal URLs:
- http://www.tandfonline.com/loi/tmph20#.VyISA1L2aic ↗
http://www.tandfonline.com/ ↗ - DOI:
- 10.1080/00268976.2022.2077258 ↗
- Languages:
- English
- ISSNs:
- 0026-8976
- Deposit Type:
- Legaldeposit
- View Content:
- Available online (eLD content is only available in our Reading Rooms) ↗
- Physical Locations:
- British Library DSC - 5900.820000
British Library DSC - BLDSS-3PM
British Library HMNTS - ELD Digital store - Ingest File:
- 22123.xml