Multi-channel convolutional neural networks for materials properties prediction

Zheng, Xiaolong; Zheng, Peng; Zheng, Liang; Zhang, Yang; Zhang, Ruizhi

doi:10.1016/j.commatsci.2019.109436

Cited by 23 publications

(8 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Naturally, such a network is not able to predict energies for mixed perovskites as it cannot distinguish between, e.g., BaTiPbO 2 and TiBaPbO 2 . To circumvent this problem, we tried two different representations along the ideas of ( 77 , 78 ) and used multiple input channels for each crystallographic position, and we ordered the input in the form of a periodic table ( Fig. 9 ).…”

Section: Methodsmentioning

confidence: 99%

Crystal graph attention networks for the prediction of stable materials

et al. 2021

View full text Add to dashboard Cite

Graph neural networks for crystal structures typically use the atomic positions and the atomic species as input. Unfortunately, this information is not available when predicting new materials, for which the precise geometrical information is unknown. We circumvent this problem by replacing the precise bond distances with embeddings of graph distances. This allows our networks to be applied directly in high-throughput studies based on both composition and crystal structure prototype without using relaxed structures as input. To train these networks, we curate a dataset of over 2 million density functional calculations of crystals with consistent calculation parameters. We apply the resulting model to the high-throughput search of 15 million tetragonal perovskites of composition ABCD 2 . As a result, we identify several thousand potentially stable compounds and demonstrate that transfer learning from the newly curated dataset reduces the required training data by 50%.

show abstract

Section: Methodsmentioning

confidence: 99%

Crystal graph attention networks for the prediction of stable materials

et al. 2021

View full text Add to dashboard Cite

show abstract

“…To this end, we rely on a well-established computational data set of Elpasolite structures reported by Faber et al 55 , which has previously served as a benchmark for regression models. [56][57][58][59][60] This set comprises the fully enumerated space of nearly two million systems, which can be derived by main group elemental exchange on the pristine quarternary Elpasolite mineral AlNaK 2 F 6 . Notably, this mineral is part of the quarternary double perovskite prototype ABC 2 D 6 , which is of significant technological interest.…”

Section: Introductionmentioning

confidence: 99%

Assessing Deep Generative Models in Chemical Composition Space

Türk

Landini

Künkel

et al. 2022

Preprint

View full text Add to dashboard Cite

The computational discovery of novel materials has been one of the main motivations behind research in theoretical chemistry for several decades. Despite much effort, this is far from a solved problem, however. Among other reasons, this is due to the enormous space of possible structures and compositions that could potentially be of interest. In the case of inorganic materials, this is exacerbated by the combinatorics of the periodic table, since even a single crystal structure can in principle display millions of compositions. Consequently, there is a need for tools that enable a more guided exploration of the materials design space. Here, generative machine learning (ML) models have recently emerged as a promising technology. In this work, we assess the performance of a range of deep generative models based on Reinforcement Learning (RL), Variational Autoencoders (VAE) and Generative Adversarial Networks (GAN) for the prototypical case of designing Elpasolite compositions with low formation energies. By relying on the fully enumerated space of 2~million main group Elpasolites, the precision, coverage and diversity of the generated materials is rigorously assessed. Additionally, a hyperparameter selection scheme for generative models in chemical composition space is developed.

show abstract

“…In recent years, convolutional neural networks (CNN) have dominated the image segmentation space by reducing the number of parameters and thus computing time and power when compared to that of a fully connected neural network [ 1 , 2 ]. Similar to multiple growing fields, materials science and engineering have benefitted from the application of CNNs in property prediction [ 3 , 4 , 5 ], the homogenization of heterogeneous materials [ 6 ], and various uses in transmission electron microscopy [ 7 , 8 , 9 ]. Surprisingly, although there have been several machine learning and semi-automated approaches applied in microstructural characterization (i.e., metallographic or ceramographic analysis) [ 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 ], the use of CNNs in grain-size determination, while growing, has been comparatively minimal.…”

Section: Introductionmentioning

confidence: 99%

Application of Deep Learning Workflow for Autonomous Grain Size Analysis

2022

View full text Add to dashboard Cite

Traditional grain size determination in materials characterization involves microscopy images and a laborious process requiring significant manual input and human expertise. In recent years, the development of computer vision (CV) has provided an alternative approach to microstructural characterization with preliminary implementations greatly simplifying the grain size determination process. Here, an end-to-end workflow to measure grain size in microscopy images without any manual input is presented. Following the ASTM standards for grain size determination, results from the line intercept (Heyn’s method) and planimetric (Saltykov’s method) approaches are used as the baseline. A pre-trained holistically nested edge detection (HED) model is used for CV-based edge detection, and the results are further compared to the classic Canny edge detection method. Post-processing was performed using open-source image processing packages to extract the grain size. In optical microscope images, the pre-trained HED model achieves much higher accuracy than the Canny edge detection method while reducing the image processing time by one to two orders of magnitude compared to traditional methods. The effects of morphological operations on the predicted grain size accuracy are also explored. Overall, the proposed end-to-end convolutional neural network (CNN)-based workflow can significantly reduce the processing time while maintaining the same accuracy as the traditional manual method.

show abstract

Multi-channel convolutional neural networks for materials properties prediction

Cited by 23 publications

References 21 publications

Crystal graph attention networks for the prediction of stable materials

Crystal graph attention networks for the prediction of stable materials

Assessing Deep Generative Models in Chemical Composition Space

Application of Deep Learning Workflow for Autonomous Grain Size Analysis

Contact Info

Product

Resources

About