Abstract
The use of machine learning methods for accelerating the design of crystalline materials usually requires manually constructed feature vectors or complex transformation of atom coordinates to input the crystal structure, which either constrains the model to certain crystal types or makes it difficult to provide chemical insights. Here, we develop a crystal graph convolutional neural networks framework to directly learn material properties from the connection of atoms in the crystal, providing a universal and interpretable representation of crystalline materials. Our method provides a highly accurate prediction of density functional theory calculated properties for eight different properties of crystals with various structure types and compositions after being trained with 10^{4} data points. Further, our framework is interpretable because one can extract the contributions from local chemical environments to global properties. Using an example of perovskites, we show how this information can be utilized to discover empirical rules for materials design.
Keywords
Computer scienceConvolutional neural networkCrystal (programming language)Representation (politics)GraphCrystal structureGraph rewritingArtificial neural networkCrystal structure predictionAtom (system on chip)Transformation (genetics)Artificial intelligenceTheoretical computer scienceAlgorithmChemistryCrystallography
Affiliated Institutions
Related Publications
Crystal engineering : the design of organic solids
1. Molecular Crystals and Crystal Engineering. Crystal engineering. Why design crystal structures of organic molecules? Some extensions. Conclusions. 2. The Atom-Atom Potential ...
Hypergraph Neural Networks
In this paper, we present a hypergraph neural networks (HGNN) framework for data representation learning, which can encode high-order data correlation in a hypergraph structure....
Heterogeneous Graph Neural Network
Representation learning in heterogeneous graphs aims to pursue a meaningful vector representation for each node so as to facilitate downstream applications such as link predicti...
Understanding image representations by measuring their equivariance and equivalence
Despite the importance of image representations such as histograms of oriented gradients and deep Convolutional Neural Networks (CNN), our theoretical understanding of them rema...
Novel tools for visualizing and exploring intermolecular interactions in molecular crystals
A new way of exploring packing modes and intermolecular interactions in molecular crystals is described, using Hirshfeld surfaces to partition crystal space. These molecular Hir...
Publication Info
- Year
- 2018
- Type
- article
- Volume
- 120
- Issue
- 14
- Pages
- 145301-145301
- Citations
- 2281
- Access
- Closed
External Links
Social Impact
Altmetric
PlumX Metrics
Social media, news, blog, policy document mentions
Citation Metrics
2281
OpenAlex
146
Influential
1926
CrossRef
Cite This
Tian Xie,
Jeffrey C. Grossman
(2018).
Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties.
Physical Review Letters
, 120
(14)
, 145301-145301.
https://doi.org/10.1103/physrevlett.120.145301
Identifiers
- DOI
- 10.1103/physrevlett.120.145301
- PMID
- 29694125
- arXiv
- 1710.10324