Machine learning has emerged as a novel tool for the efficient prediction of materials properties, and claims have been made that machine-learned models for the formation energy of compounds can approach the accuracy of Density Functional Theory (DFT). The models tested in this work include five recently published compositional models, a baseline model using stoichiometry alone, and a structural model. By testing seven machine learning models for formation energy on stability predictions using the Materials Project database of DFT calculations for 85,014 unique chemical compositions, we show that while formation energies can indeed be predicted well, all compositional models perform poorly on predicting the stability of compounds, making them considerably less useful than DFT for the discovery and design of new solids. Most critically, in sparse chemical spaces where few stoichiometries have stable compounds, only the structural model is capable of efficiently detecting which materials are stable. The non-incremental improvement of structural models compared with compositional models is noteworthy and encourages the use of structural models for materials discovery, with the constraint that for any new composition, the ground-state structure is not known a priori. This work demonstrates that accurate predictions of formation energy do not imply accurate predictions of stability, emphasizing the importance of assessing model performance on stability predictions, for which we provide a set of publicly available tests.
Prediction of low-Z collinear and noncollinear antiferromagnetic compounds having momentum-dependent spin splitting even without spin-orbit coupling
最近的研究（袁等。例如，Phys. Rev. B 102，014422（2020））揭示了由反铁磁有序诱导的SOC独立的自旋分裂和自旋极化效应，其不一定需要反转对称性的破坏或SOC的存在，因此甚至可以存在于中心对称的低Z轻元素化合物中，从而大大拓宽了自旋极化的材料基础。在目前的工作中，我们开发的磁对称性条件，使这种效果，分为7个不同的自旋分裂原型（SST-1到SST-7）的1651磁空间群。我们使用“逆向设计”的方法，首先制定的目标属性（在这里，自旋分裂低Z化合物不限于低对称结构），然后推导出使物理设计原则，以搜索可实现的化合物，满足这些先验的设计原则。这一过程揭示了422个磁空间群（160个中心对称和262个非中心对称），这些磁空间群可以保持AFM诱导的、与SOC无关的自旋分裂和自旋极化。然后，我们寻找稳定的化合物，使这种对称性。我们采用密度泛函理论（DFT）研究了一些典型化合物的电子和自旋结构，发现了不同于传统Rashba-Dresselhaus模式的自旋织构.我们提供了所有反铁磁自旋分裂原型（SST-1到SST-4）的DFT结果，并集中于揭示AFM诱导的自旋分裂原型（SST-4）。对称性设计原理及其转化为逆向设计的材料搜索方法和DFT验证可以为实验检验开辟道路。对称性设计原理及其转化为逆向设计材料搜索方法和DFT验证可以为其实验检查开辟道路。
Accurate Screening of Functional Materials with Machine-Learning Potential and Transfer-Learned Regressions: Heusler Alloy Benchmark
Research Center for Magnetic and Spintronic Materials, National Institute for Materials Science（磁性及自旋材料研究中心，物质材料研究机构）；Digital Transformation Initiative Center for Magnetic Materials (DXMag), National Institute for Materials Science（磁性材料数字化转型倡议中心（DXMag），物质材料研究机构）
Evaluating Mechanical Property Prediction across Material Classes using Molecular Dynamics Simulations with Universal Machine-Learned Interatomic Potentials
Large Scale Structure Prediction of Near-Stoichiometric Magnesium Oxide Based on a Machine-Learned Interatomic Potential: Novel Crystalline Phases and Oxygen-Vacancy Ordering
利用快速准确的神经网络势，我们能够用最小跳跃方法系统地探索大块氧化镁的大晶胞的能量景观。该潜力的训练重点是近化学计量组成，特别是低氧化物，即，Mg $_x$ O $_{1-x}$与$0.50<x<0.60$。我们广泛的探索表明，对于散装化学计量的化合物，有几个新的低能量岩盐状结构，其中Mg原子是八面体六配位的，并形成具有不同堆叠顺序的三角棱柱图案。此外，对于$x$的每种组成，我们发现了Mg $_x$ O $_{1-x}$的新型非化学计量晶相的密集光谱。这些结构大多类似于具有八面体配位和五配位Mg原子的岩盐结构。由于去除了一个氧原子，能量景观变得更像玻璃，具有氧空位型结构，它们在能量上彼此非常接近。对于相同数量的镁和氧原子，如果空位沿着线或平面排列，则我们的缺氧结构的能量低于具有随机分布的氧空位的岩盐结构。我们还发现了假定的全球最低配置的非化学计量的低氧化物结构的每个组合物。这些结构主要由岩盐结构的（111）板片组成，其在顶部和底部用Mg原子终止，并且沿着$z$方向以不同的顺序堆叠。与其他Magnéli型相一样，这些结构具有与其化学计量对应物显著不同的性质，例如低晶格热导率和高电导率。
Constructing accurate machine-learned potentials and performing highly efficient atomistic simulations to predict structural and thermal properties
School of Engineering and Applied science, University of Pennsylvania（宾夕法尼亚大学工程与应用科学学院）；College of Resources and Environment, Chengdu University of Information Technology（成都信息工程大学资源环境学院）；USC Keck School of Medicine, University of Southern California（南加州大学凯克医学院）；College of Big Data and Intelligent Engineering, Southwest Forestry University（西南林业大学大数据与智能工程学院）；College of Information Sciences and Technology, Donghua University（东华大学信息科学与技术学院）
