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 materials.It 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 set.Traditional 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.
基金
supported in part by National Science Foundation under the grants 2110033,OAC-2311203,and 2320292.
