Formation energy prediction of crystalline compounds using deep convolutional network learning on voxel image representation

Formation energy prediction of crystalline compounds using deep convolutional network learning on voxel image representation
复制标题

使用体素图像表示的深度卷积网络学习来预测晶体化合物的形成能

DOI:
10.1038/s43246-023-00433-9
复制
发表时间:
2023
影响因子:
7.8
通讯作者:
Kadkhodaei, Sara
Kadkhodaei, Sara
中科院分区:
--
文献类型:
--
作者:
Davariashtiyani, Ali;Kadkhodaei, Sara

文献摘要

参考文献

相似文献

新兴的机器学习模型已经能够有效和准确地预测化合物的形成能量,最流行的模型依赖于图形结构来表示晶体材料。在这里,我们介绍了一种替代方法的基础上稀疏体素图像的晶体。通过开发一个复杂的网络架构,我们展示了从视觉图像表示中学习无机化合物结构和化学排列的基本特征的能力,随后将这些特征与化合物的形成能相关联。我们的模型通过在深度卷积网络中利用跳过连接并在训练过程中加入旋转晶体样本的增强来实现准确的地层能量预测,其性能与最先进的方法相当。通过采用视觉图像作为晶体化合物的替代表示,并利用深度卷积网络的能力,这项研究扩展了机器学习的前沿,以加速材料发现和优化。在一个全面的评估,我们分析了3115二元系统的预测凸包,并介绍了错误的形成能量误差以外的度量。这种评估提供了有价值的见解的影响,形成能量误差的预测凸壳的性能。
Emerging machine-learned models have enabled efficient and accurate prediction of compound formation energy, with the most prevalent models relying on graph structures for representing crystalline materials. Here, we introduce an alternative approach based on sparse voxel images of crystals. By developing a sophisticated network architecture, we showcase the ability to learn the underlying features of structural and chemical arrangements in inorganic compounds from visual image representations, subsequently correlating these features with the compounds’ formation energy. Our model achieves accurate formation energy prediction by utilizing skip connections in a deep convolutional network and incorporating augmentation of rotated crystal samples during training, performing on par with state-of-the-art methods. By adopting visual images as an alternative representation for crystal compounds and harnessing the capabilities of deep convolutional networks, this study extends the frontier of machine learning for accelerated materials discovery and optimization. In a comprehensive evaluation, we analyse the predicted convex hulls for 3115 binary systems and introduce error metrics beyond formation energy error. This evaluation offers valuable insights into the impact of formation energy error on the performance of the predicted convex hulls.
DOI: 10.1088/2515-7655/abe425
发表时间: 2021-04
期刊: Journal of Physics: Energy
影响因子: --
作者:
Gordon G C Peterson;Jakoah Brgoch
通讯作者: Gordon G C Peterson;Jakoah Brgoch
DOI: 10.1038/s41524-021-00650-1
发表时间: 2021-11-15
影响因子: 9.7
作者:
Choudhary, Kamal;DeCost, Brian
通讯作者: DeCost, Brian
DOI: 10.1063/1.4812323
发表时间: 2013-07-01
期刊: APL MATERIALS
影响因子: 6.1
作者:
Jain, Anubhav;Shyue Ping Ong;Persson, Kristin A.
通讯作者: Persson, Kristin A.
DOI: 10.1088/2632-2153/acb315
发表时间: 2023
期刊: Machine Learning: Science and Technology
影响因子: --
作者:
Yongqiang Cheng;Geoffrey Wu;D. Pajerowski;M. Stone;A. Savici;Mingda Li;A. Ramirez‐Cuesta
通讯作者: A. Ramirez‐Cuesta
DOI: 10.1038/s43246-021-00219-x
发表时间: 2021-11-18
影响因子: 7.8
作者:
Davariashtiyani, Ali;Kadkhodaie, Zahra;Kadkhodaei, Sara
通讯作者: Kadkhodaei, Sara