该研究用 FLARE 与 NequIP 对比 TM23 中过渡金属的力、能量和应力预测，揭示早期过渡金属因费米能级附近尖锐 d 态密度而具有更复杂势能面，并提供基准数据集。
摘要：
本工作考察了机器学习力场（MLFF）在描述 d 区元素块体固相和液相时与准确性相关的挑战。我们极其详尽地比较了两种领先 MLFF 模型在各过渡金属上的力、能量和应力预测性能：一种是使用稀疏高斯过程实现的、基于核的原子簇展开方法（FLARE），另一种是等变消息传递神经网络（NequIP）。与晚期铂族和货币族元素相比，早期过渡金属表现出更高的相对误差，也更难学习，而且这一趋势在不同模型架构中均持续存在。通过比较具有不同多体阶数和角分辨率的表示的性能，揭示了不同金属原子间相互作用复杂性的趋势。利用基于费米能级附近占据和未占据 d 态的微扰理论论证，我们确定早期过渡金属在费米能级上下均存在巨大而尖锐的 d 态密度，这导致这些金属具有更复杂、更难学习的势能面。提高虚拟电子温度（展宽）会改变力的角度敏感性，并使早期过渡金属的力更容易学习。本工作阐明了使用当前领先 MLFF 捕捉金属键复杂性质所面临的挑战，并提供了一个过渡金属参考数据集，旨在对新兴机器学习近似方法进行准确性基准测试并促进其发展。
英文摘要：
This work examines challenges associated with the accuracy of machine-learned force fields (MLFFs) for bulk solid and liquid phases of d-block elements. In exhaustive detail, we contrast the performance of force, energy, and stress predictions across the transition metals for two leading MLFF models: a kernel-based atomic cluster expansion method implemented using sparse Gaussian processes (FLARE), and an equivariant message-passing neural network (NequIP). Early transition metals present higher relative errors and are more difficult to learn relative to late platinum- and coinage-group elements, and this trend persists across model architectures. Trends in complexity of interatomic interactions for different metals are revealed via comparison of the performance of representations with different many-body order and angular resolution. Using arguments based on perturbation theory on the occupied and unoccupied d states near the Fermi level, we determine that the large, sharp d density of states both above and below the Fermi level in early transition metals leads to a more complex, harder-to-learn potential energy surface for these metals. Increasing the fictitious electronic temperature (smearing) modifies the angular sensitivity of forces and makes the early transition metal forces easier to learn. This work illustrates challenges in capturing intricate properties of metallic bonding with current leading MLFFs and provides a reference data set for transition metals, aimed at benchmarking the accuracy and improving the development of emerging machine-learned approximations.
From Energy-Force Weighting to Primal-Dual Optimization of Machine-Learned Interatomic Potentials
School of Mathematical Sciences, Institute of Natural Sciences and MOE-LSC, Shanghai Jiao Tong University（上海交通大学数学科学学院、自然科学研究院和教育部线性科学中心）；Department of Mathematics, National University of Singapore（新加坡国立大学数学系）
Machine-Learned Force Fields for Lattice Dynamics at Coupled-Cluster Level Accuracy
Institute of Theoretical Physics, Technical University of Vienna（维也纳技术大学理论物理研究所）；Faculty of Physics and Center for Computational Materials Science, University of Vienna（维也纳大学物理系和计算材料科学中心）
Electrostatic Phenomenology Benchmarks for Machine-Learned Interatomic Potentials in Electrochemistry: Beyond the Energy-Force Metric
Fritz-Haber-Institut der Max-Planck-Gesellschaft（马克斯·普朗克学会弗里茨·哈伯研究所）；Department of Chemistry and Biochemistry, University of South Carolina（南卡罗来纳大学化学与生物化学系）；Department of Chemical Engineering, University of California, Davis（加州大学戴维斯分校化学工程系）；Department of Materials, Imperial College London（帝国理工学院材料系）；Department of Chemistry, Imperial College London（帝国理工学院化学系）；Department of Materials Science and Engineering, University of Florida（佛罗里达大学材料科学与工程系）；Department of Chemistry, Korea University（高丽大学化学系）；Max Planck Institute for Sustainable Materials（马克斯·普朗克可持续材料研究所）
Developing Machine-Learned Potentials for Coarse-Grained Molecular Simulations: Challenges and Pitfalls
Institute of Informatics and Telecommunications（信息学与电信研究所）；Institute of Nanoscience and Nanotechnology（纳米科学与纳米技术研究所）；National Centre for Scientific Research “Demokritos”（“德谟克利特”国家科学研究中心）；SciFY P.N.P.C.（SciFY公共非营利公司）；School of Chemical Engineering（化学工程学院）；National Technical University of Athens（雅典国立技术大学）
Inverse design for Casimir-Lifshitz force near heterogeneous gapped metal surface
Centre of Excellence ENSEMBLE3 Sp. z o. o.（ENSEMBLE3卓越中心有限责任公司）；Chemical and Biological Systems Simulation Lab, Centre of New Technologies, University of Warsaw（华沙大学新技术中心化学与生物系统模拟实验室）；Department of Physics, Ferdowsi University of Mashhad（马什哈德费尔多西大学物理系）；Zernike Institute for Advanced Materials, University of Groningen（格罗宁根大学泽尔尼克先进材料研究所）
Electromotive forces generated in 3d-transition ferromagnetic metal films themselves under their ferromagnetic resonance
报道了3d过渡铁磁金属（FM = Fe，Co，Ni_（80）Fe_（20））薄膜在铁磁共振（FMR）作用下产生的电动势（EMF）特性。对于Fe和Co薄膜，由于反常霍尔效应产生的电动势主要是在它们的铁磁电阻下产生的。同时，对于Ni 80 Fe 20薄膜，主要是由于FMR下Ni 80 Fe 20薄膜本身的逆自旋霍尔效应产生的电动势。这种趋势定性地解释了不同的自旋极化，自旋霍尔电导率，反常霍尔电导率，磁化饱和，和FM膜的电阻率。
Machine-Learned Interatomic Potentials for Predicting Physicochemical Properties of Molten Metal-Salt Systems for Calcium Electrolysis
Skolkovo Institute of Science and Technology（斯科尔科沃科学技术学院）；Digital Materials LLC（数字材料有限责任公司）；Institute of High Temperature Electrochemistry, Ural Branch of the Russian Academy of Sciences（俄罗斯科学院乌拉尔分院高温电化学研究所）
Observing Nucleation and Crystallization of Rocksalt LiF from Molten State through Molecular Dynamics Simulations with Refined Machine-Learned Force Field
Synthesizability, hardness, and stacking order in multicomponent transition metal carbides from machine-learned potentials
École Polytechnique Fédérale de Lausanne（洛桑联邦理工学院）；Institute of Materials, École Polytechnique Fédérale de Lausanne（洛桑联邦理工学院材料研究所）；Laboratory of materials design and simulation (MADES), Institute of Materials, École Polytechnique Fédérale de Lausanne（洛桑联邦理工学院材料研究所材料设计与模拟实验室）；National Centre for Computational Design and Discovery of Novel Materials (MARVEL), École Polytechnique Fédérale de Lausanne（洛桑联邦理工学院新材料计算设计与发现国家中心）
Why the Casimir Force for Magnetic Metals Computed by the Lifshitz Theory Using the Drude Model Disagrees with the Measurement Data
Central Astronomical Observatory at Pulkovo of the Russian Academy of Sciences（俄罗斯科学院普尔科沃中央天文台）；Peter the Great Saint Petersburg Polytechnic University（彼得大帝圣彼得堡理工大学）；Moscow Center for Advanced Studies（莫斯科高等研究院）
