Structure-based out-of-distribution(OOD)materials property prediction:a benchmark study

作     者：Sadman Sadeed Omee Nihang Fu Rongzhi Dong Ming Hu Jianjun Hu 

作者机构：Department of Computer Science and EngineeringUniversity of South CarolinaColumbiaSCUSA Department of Mechanical EngineeringUniversity of South CarolinaColumbiaSCUSA 

出 版 物：《npj Computational Materials》 (计算材料学(英文))

年 卷 期：2024年第10卷第1期

页      面：1753-1766页

核心收录：

学科分类：08[工学] 080502[工学-材料学] 0805[工学-材料科学与工程（可授工学、理学学位）] 

基　　金：supported in part by National Science Foundation under the grants 2110033 OAC-2311203 and 2320292 

主　　题：property prediction distribution 

摘      要：In real-world materials research,machine learning(ML)models are usually expected to predict and discover novel exceptional materials that deviate from the known *** is thus a pressing question to provide an objective evaluation ofMLmodel performances in property prediction of out-ofdistribution(OOD)materials that are different fromthe training *** performance evaluation of materials property prediction models through the random splitting of the dataset frequently results in artificially high-performance assessments due to the inherent redundancy of typical material datasets.