论文标题
预测具有原子线图神经网络（Alignn）的晶体化合物的电子密度
Prediction of the electron density of states for crystalline compounds with Atomistic Line Graph Neural Networks (ALIGNN)
论文作者
论文摘要
基于机器学习（ML）的模型大大增强了传统的材料发现和设计管道。具体而言，近年来，用于材料属性预测的替代ML模型在预测离散标量值目标属性方面取得了成功，以在其DFT计算值的合理准确性之内。但是，由于目标的复杂性和有限的可用训练数据，因此准确预测频谱目标（例如状态的电子密度（DOS））提出了更具挑战性的问题。在这项研究中，我们介绍了最近开发的原子图神经网络（Alignn）的扩展，以准确预测一套大量材料单位细胞结构的DO，该结构对公开可用的Jarvis-DFT数据集进行了训练。此外，我们评估了目标数量的两种表示方法 - 直接离散的频谱以及使用自动编码器获得的压缩的低维表示。通过这项工作，我们证明了基于图形的特征和建模方法的实用性，以预测依赖材料结构的化学和方向特征的复杂靶标。
Machine learning (ML) based models have greatly enhanced the traditional materials discovery and design pipeline. Specifically, in recent years, surrogate ML models for material property prediction have demonstrated success in predicting discrete scalar-valued target properties to within reasonable accuracy of their DFT-computed values. However, accurate prediction of spectral targets such as the electron Density of States (DOS) poses a much more challenging problem due to the complexity of the target, and the limited amount of available training data. In this study, we present an extension of the recently developed Atomistic Line Graph Neural Network (ALIGNN) to accurately predict DOS of a large set of material unit cell structures, trained to the publicly available JARVIS-DFT dataset. Furthermore, we evaluate two methods of representation of the target quantity - a direct discretized spectrum, and a compressed low-dimensional representation obtained using an autoencoder. Through this work, we demonstrate the utility of graph-based featurization and modeling methods in the prediction of complex targets that depend on both chemistry and directional characteristics of material structures.
