Small data machine learning in materials science

• Published in: npj computational materials Vol. 9; no. 1; pp. 42 - 15
• Main Authors: Xu, Pengcheng, Ji, Xiaobo, Li, Minjie, Lu, Wencong
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 25.03.2023; Nature Publishing Group; Nature Portfolio
• Subjects: 639/301/1034; 639/301/1034/1037; Active learning; Algorithms; Artificial intelligence; Big Data; Characterization and Evaluation of Materials; Chemistry and Materials Science; Computational Intelligence; Data collection; Experiments; Genomes; Interdisciplinary subjects; Learning algorithms; Machine learning; Materials Science; Mathematical and Computational Engineering; Mathematical and Computational Physics; Mathematical Modeling and Industrial Mathematics; Review Article; Theoretical; Transfer learning; Workflow
• ISSN: 2057-3960; 2057-3960
• DOI: 10.1038/s41524-023-01000-z

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.