Computational Screening of New Perovskite Materials Using Transfer Learning and Deep Learning
Year: 2019
Type: article
Abstract: As one of the most studied materials, perovskites exhibit a wealth of superior properties that lead to diverse applications. Computational prediction of novel stable perovskite structures has big potential in the discovery of new materials for solar panels, superconductors, thermal electric, and catalytic materials, etc. By addressing one of the key obstacles of machine learning based materials discovery, the lack of sufficient training data, this paper proposes a transfer learning based approach that exploits the high accuracy of the machine learning model trained with physics-informed structural and elemental descriptors. This gradient boosting regressor model (the transfer learning model) allows us to predict the formation energy with sufficient precision of a large number of materials of which only the structural information is available. The enlarged training set is then used to train a convolutional neural network model (the screening model) with the generic Magpie elemental features with high prediction power. Extensive experiments demonstrate the superior performance of our transfer learning model and screening model compared to the baseline models. We then applied the screening model to filter out promising new perovskite materials out of 21,316 hypothetical perovskite structures with a large portion of them confirmed by existing literature.
Source: Applied Sciences
DOI: 10.3390/app9245510
Authors: Xiang Li(Conceptualization, Investigation, Methodology, Software, Validation, Writing – original draft, Writing – review & editing), Yabo Dan(Investigation, Methodology, Software), Rongzhi Dong(Investigation), Zhuo Cao(Investigation), Chengcheng Niu(Validation), Yuqi Song(Investigation), Shaobo Li(Funding acquisition, Resources, Supervision, Writing – review & editing), Jianjun Hu(Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing – original draft, Writing – review & editing)
Institutions: Guizhou University, University of South Carolina
Language: English
Subfield: Materials Chemistry
Field: Materials Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $1,520
Awards: G2941082053, Collaborative Research: Integrating Physics and Generative Machine Learning Models for Inverse Materials Design, RII Track 1: Materials Assembly and Design Excellence in South Carolina: MADE in SC, G5225387144, G615300131, G8776736351, EAGER: Thermal Materials Discovery via Deep Learning based High-Throughput Computational Screening
OpenAlex ID: W2994650566
