Abstract

This review discussed the dilemma of small data faced by materials machine learning. First, we analyzed the limitations brought by small data. Then, the workflow of materials machine learning has been introduced. Next, the methods of dealing with small data were introduced, including data extraction from publications, materials database construction, high-throughput computations and experiments from the data source level; modeling algorithms for small data and imbalanced learning from the algorithm level; active learning and transfer learning from the machine learning strategy level. Finally, the future directions for small data machine learning in materials science were proposed.

Details

Title
Small data machine learning in materials science
Author
Xu, Pengcheng 1 ; Ji, Xiaobo 2 ; Li, Minjie 2   VIAFID ORCID Logo  ; Lu, Wencong 3   VIAFID ORCID Logo 

1  Shanghai University, Materials Genome Institute, Shanghai, China (GRID:grid.39436.3b) (ISNI:0000 0001 2323 5732) 
2  Shanghai University, Department of Chemistry, College of Sciences, Shanghai, China (GRID:grid.39436.3b) (ISNI:0000 0001 2323 5732) 
3  Shanghai University, Materials Genome Institute, Shanghai, China (GRID:grid.39436.3b) (ISNI:0000 0001 2323 5732); Shanghai University, Department of Chemistry, College of Sciences, Shanghai, China (GRID:grid.39436.3b) (ISNI:0000 0001 2323 5732); Zhejiang Laboratory, Hangzhou, China (GRID:grid.510538.a) (ISNI:0000 0004 8156 0818) 
Pages
42
Publication year
2023
Publication date
2023
Publisher
Nature Publishing Group
e-ISSN
20573960
Source type
Scholarly Journal
Language of publication
English
DOI
https://doi.org/10.1038/s41524-023-01000-z
ProQuest document ID
2790694538
Copyright
© The Author(s) 2023. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.