PROFiT-Net: Property-Networking Deep Learning Model for Materials,Journal of the American Chemical Society

当前位置： X-MOL 学术 › J. Am. Chem. Soc. › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

PROFiT-Net: Property-Networking Deep Learning Model for Materials
Journal of the American Chemical Society ( IF 14.4 ) Pub Date : 2024-09-12 , DOI: 10.1021/jacs.4c05159
Se-Jun Kim ₁ , Won June Kim ₂ , Changho Kim ₃ , Eok Kyun Lee ₁ , Hyungjun Kim ₁

Affiliation

There is a growing need to develop artificial intelligence technologies capable of accurately predicting the properties of materials. This necessitates the expansion of material databases beyond the scope of density functional theory, and also the development of deep learning (DL) models that can be effectively trained with a limited amount of high-fidelity data. We developed a DL model utilizing a crystal structure representation based on the orbital field matrix (OFM), which was modified to incorporate information on elemental properties and valence electron configurations. This model, effectively capturing the interrelation between the elemental properties in the crystal, was coined the PRoperty-networking Orbital Field maTrix-convolutional neural Network (PROFiT-Net). Remarkably, PROFiT-Net demonstrated high accuracy in predicting the dielectric constant, experimental band gaps, and formation enthalpies compared with other leading DL models. Moreover, our model accurately identifies physical patterns, such as avoiding the prediction of unphysical negative band gaps and exhibiting a Penn-model-like trend while maintaining the scalability. We envision that PROFiT-Net will accelerate the development of functional materials.

中文翻译：

PROFiT-Net：材料的属性网络深度学习模型

人们越来越需要开发能够准确预测材料特性的人工智能技术。这就需要将材料数据库扩展到密度泛函理论的范围之外，并且需要开发可以使用有限数量的高保真数据进行有效训练的深度学习（DL）模型。我们开发了一个利用基于轨道场矩阵 (OFM) 的晶体结构表示的深度学习模型，该模型经过修改以纳入元素特性和价电子构型的信息。该模型有效地捕捉了晶体中元素属性之间的相互关系，被称为属性网络轨道场矩阵卷积神经网络（PROFiT-Net）。值得注意的是，与其他领先的深度学习模型相比，PROFiT-Net 在预测介电常数、实验带隙和形成焓方面表现出较高的准确性。此外，我们的模型准确地识别了物理模式，例如避免了非物理负带隙的预测并在保持可扩展性的同时表现出类似佩恩模型的趋势。我们预计PROFiT-Net将加速功能材料的开发。

更新日期：2024-09-12

点击分享查看原文

点击收藏

阅读更多本刊最新论文本刊介绍/投稿指南11