AtomSets as a hierarchical transfer learning framework for small and large materials datasets

Journal Article · 22 October 2021 · npj Computational Materials

DOI: https://doi.org/10.1038/s41524-021-00639-w · OSTI ID:1827232

Chen, Chi ; Ong, Shyue Ping

Abstract Predicting properties from a material’s composition or structure is of great interest for materials design. Deep learning has recently garnered considerable interest in materials predictive tasks with low model errors when dealing with large materials data. However, deep learning models suffer in the small data regime that is common in materials science. Here we develop the AtomSets framework, which utilizes universal compositional and structural descriptors extracted from pre-trained graph network deep learning models with standard multi-layer perceptrons to achieve consistently high model accuracy for both small compositional data (<400) and large structural data (>130,000). The AtomSets models show lower errors than the graph network models at small data limits and other non-deep-learning models at large data limits. They also transfer better in a simulated materials discovery process where the targeted materials have property values out of the training data limits. The models require minimal domain knowledge inputs and are free from feature engineering. The presented AtomSets model framework can potentially accelerate machine learning-assisted materials design and discovery with less data restriction.

• Sponsoring Organization: USDOE
• Grant/Contract Number: NONE; AC02-05CH11231
• OSTI ID: 1827232
• Journal Information: npj Computational Materials, Journal Name: npj Computational Materials Journal Issue: 1 Vol. 7; ISSN 2057-3960
• Publisher: Nature Publishing GroupCopyright Statement
• Country of Publication: United Kingdom
• Language: English