Systematic assessment of various universal machine-learning interatomic potentials  

作　　者：Haochen Yu Matteo Giantomassi Giuliana Materzanini Junjie Wang Gian-Marco Rignanese 

机构地区：[1]Institute of Condensed Matter and Nanosciences,Universitécatholique de Louvain,Louvain-la-Neuve,Belgium [2]State Key Laboratory of Solidification Processing,Northwestern Polytechnical University,Xi'an,Shaanxi,China [3]WEL Research Institute,Wavre,Belgium

出　　处：《Materials Genome Engineering Advances》2024年第3期59-70,共12页材料基因工程前沿（英文）

基　　金：supported by the National Key Research and Development Program of China(2022YFE0141100 and 2023YFB3003005).

摘　　要：Machine-learning interatomic potentials have revolutionized materials modeling at the atomic scale.Thanks to these,it is now indeed possible to perform simulations of ab initio quality over very large time and length scales.More recently,various universal machine-learning models have been proposed as an out-of-box approach avoiding the need to train and validate specific potentials for each particular material of interest.In this paper,we review and evaluate four different universal machine-learning interatomic potentials(uMLIPs),all based on graph neural network architectures which have demonstrated transferability from one chemical system to another.The evaluation procedure relies on data both from a recent verification study of density-functional-theory implementations and from the Materials Project.Through this comprehensive evaluation,we aim to provide guidance to materials scientists in selecting suitable models for their specific research problems,offer recommendations for model selection and optimization,and stimulate discussion on potential areas for improvement in current machinelearning methodologies in materials science.

关 键 词：formation energy geometry optimization machine learning phonons universal machine-learning interatomic potentials VERIFICATION 

分 类 号：TB3[一般工业技术—材料科学与工程] TP181[自动化与计算机技术—控制理论与控制工程]