A geometric-information-enhanced crystal graph network for predicting properties of materials

Cheng, Jiucheng; Zhang, Chunkai; Dong, Lifeng

doi:10.1038/s43246-021-00194-3

Cited by 57 publications

(51 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After that, the CGCNN model was further improved. The improved variant of the CGCNN model [101] and the geometry-information-enhanced crystal GNN [58] appeared. The former connects the clear three-body association of adjacent constituent atoms and optimizes the chemical representation of atomic bonds in the crystal diagram, which shortens the high-throughput search time and achieves superiority to CGCNN.…”

Section: Graph Neural Networkmentioning

confidence: 99%

“…Since the accuracy of DL models will significantly affect their potency in material science, a natural way to improve model accuracy is to use more training data. However, it should be noted that suitable model architecture must be adopted to guarantee that the characteristics of the material data are full exploited [58]. By combining with other traditional optimization algorithms, DL can also benefit for better accuracy [59].…”

mentioning

confidence: 99%

See 1 more Smart Citation

Deep learning modeling strategy for material science: from natural materials to metamaterials

Chen

Xiong

et al. 2022

J. Phys. Mater.

View full text Add to dashboard Cite

Computational modeling is a crucial approach in material-related research for discovering new materials with superior properties. However, the high design flexibility in materials, especially in the realm of metamaterials where the sub-wavelength structure provides an additional degree of freedom in design, poses a formidable computational cost in various real-world applications. With the advent of big data, deep learning brings revolutionary breakthroughs in many conventional machine learning and pattern recognition tasks such as image classification. The accompanied data-driven modeling paradigm also provides transformative methodology shift in materials science, from trial-and-error routine to intelligent material discovery and analysis. This review systematically summarize the application of deep learning in material science, based on a model selection perspective for both natural materials and metamaterials. The review aims to uncover the logic behind data-model relation with emphasis on suitable data structures for different scenarios in the material study and the corresponding problem-solving deep learning model architectures.

show abstract

Section: Graph Neural Networkmentioning

confidence: 99%

mentioning

confidence: 99%

Deep learning modeling strategy for material science: from natural materials to metamaterials

Chen

Xiong

et al. 2022

J. Phys. Mater.

View full text Add to dashboard Cite

show abstract

“…This enables the model to learn the best fea- tures to represent the structure as opposed to the typical "handcrafted" feature approach [24] and achieve a formation energy validation MAE of 39 meV/atom [23]. More recently, the MAE of formation energy prediction of graph-based models continued to decrease to 21-39 meV/atom [25][26][27][28][29].…”

Section: Introductionmentioning

confidence: 99%

Data-Augmentation for Graph Neural Network Learning of the Relaxed Energies of Unrelaxed Structures

Gibson¹,

Hire²,

Hennig³

2022

Preprint

View full text Add to dashboard Cite

Computational materials discovery has continually grown in utility over the past decade due to advances in computing power and crystal structure prediction algorithms (CSPA). However, the computational cost of the ab initio calculations required by CSPA limits its utility to small unit cells, reducing the compositional and structural space the algorithms can explore. Past studies have bypassed many unneeded ab initio calculations by utilizing machine learning methods to predict formation energy and determine the stability of a material. Specifically, graph neural networks display high fidelity in predicting formation energy. Traditionally graph neural networks are trained on large data sets of relaxed structures. Unfortunately, the geometries of unrelaxed candidate structures produced by CSPA often deviate from the relaxed state, which leads to poor predictions hindering the model's ability to filter energetically unfavorable prior to ab initio evaluation. This work shows that the prediction error on relaxed structures reduces as training progresses, while the prediction error on unrelaxed structures increases, suggesting an inverse correlation between relaxed and unrelaxed structure prediction accuracy. To remedy this behavior, we propose a simple, physically motivated, computationally cheap perturbation technique that augments training data to improve predictions on unrelaxed structures dramatically. On our test set consisting of 623 Nb-Sr-H hydride structures, we found that training a crystal graph convolutional neural networks, utilizing our augmentation method, reduced the MAE of formation energy prediction by 66% compared to training with only relaxed structures. We then show how this error reduction can accelerates CSPA by improving the model's ability to filter out energetically unfavorable structures accurately.

show abstract

“…Choudhary et al [10] proposed an Atomistic Line Graph Neural Network (ALIGNN), which constructs the bond graph between atoms, and the bond angle information can be transmitted between atoms through the graph neural network. Cheng et al [11] proposed a Geometric Information Enhanced Crystal Network (GeoCGNN) to predict material properties. GeoCGNN designs a mixed basis function which includes Gaussian radial basis and plane wave to encode the geometric structure information.…”

Section: Introductionmentioning

confidence: 99%

TG-GNN: Transformer based Geometric enhancement graph neural network for molecular property prediction

Zhang

Guo²,

Chen

et al. 2022

Preprint

View full text Add to dashboard Cite

Molecular properties play a crucial role in material discovery, protein interaction and drug development. The appearance of Graph Neural Network(GNN) significantly improved the performance of molecular property prediction. However, nodes in GNN only update the features of neighbor nodes, resulting in insufficient ability to encode global feature information. The self attention mechanism in transformer encodes the global information of molecules, while its spatial information is insufficient. Since molecules are three-dimensional spatial structures, spatial geometry information is an important attribute of molecules. Therefore, a transformer based geometric enhanced graph neural network (TG-GNN) is proposed to combine global and local molecule information with three-dimensional spatial structures to predict molecular properties. In this network, Graph neural network Geometric Feature Fusion Module (GGFF) and Transformer Geometric Feature Enhancement Module (TGF) are proposed to enhance the spatial geometry learning ability of the network. The GGFF module constructs a parallel graph neural network using bond and bond angle. The introduction of bond angle effectively complements the spatial information of the network. The TGF module introduces the coordinates and centrality degree features into the transformer to enhance the geometric expression ability of the module. GGFF module and TGF module encode local information and global information of molecules at the same time on the QM9 and OMDB dataset.

show abstract

A geometric-information-enhanced crystal graph network for predicting properties of materials

Cited by 57 publications

References 41 publications

Deep learning modeling strategy for material science: from natural materials to metamaterials

Deep learning modeling strategy for material science: from natural materials to metamaterials

Data-Augmentation for Graph Neural Network Learning of the Relaxed Energies of Unrelaxed Structures

TG-GNN: Transformer based Geometric enhancement graph neural network for molecular property prediction

Contact Info

Product

Resources

About