Rational Design of Deep Learning Networks Based on a Fusion Strategy for Improved Material Property Predictions
Journal of Chemical Theory and Computation · 13.08.2024 · 20 (15) · 6756-6771
DOI: 10.1021/acs.jctc.4c00187 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Chemical Theory and Computation, online erschienen am 18.07.2024. Verfasst von Hongwei Du und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 59 Literaturangaben. Gefördert von: National Key Research and Development Program of China.
Abstract (Originalsprache)
The success of machine learning in predicting material properties is largely dependent on the design of the model. However, the current designs of deep learning models in materials science have the following prominent problems. First, the model design lacks a rational guidance strategy and heavily relies on a large amount of trial and error. Second, numerous deep learning models are utilized across various fields, each with its own advantages and disadvantages. Therefore, it is important to incorporate a fusion strategy to fully leverage them and further expand the design strategies of the models. To address these problems, we analyze that the main reason is the lack of a new feedback method rich in physical insights. In this study, we developed a feedback method called the Chemical Environment Clustering Vector (CECV) of compounds at different thresholds, which is rich in physical insights. Based on CECV, we rationally designed the Long Short-Term Memory and Gated Recurrent Unit fused with Deep Convolutional Neural Network (L-G-DCNN) to explore the field of structure-agnostic material property predictions. L-G-DCNN accurately captures the interactions between elements in compounds, enabling more accurate and efficient predictions of the material properties. Our results demonstrate that the performance of the L-G-DCNN surpasses the current state-of-the-art structure-agnostic models across 28 benchmark data sets, exhibiting superior sample efficiency and faster convergence speed. By employing different visualization methods, we demonstrate that the fusion strategy based on CECV significantly enhances the comprehension of the L-G-DCNN model design and provides a fresh perspective for researchers in the field of materials informatics.
Autor:innen und Institutionen
- School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China.
- Zhangjiang Institute for Advanced Study, Shanghai Jiao Tong University, Shanghai 201203, China.
- Materials Genome Initiative Center, Shanghai Jiao Tong University, Shanghai 200240, China.
ORCID-Profile: Lanting Zhang, Hong Wang
Förderung
- National Key Research and Development Program of China
Zitieren
APA
Du, H., Hui, J., Zhang, L., & Wang, H. (2024). Rational Design of Deep Learning Networks Based on a Fusion Strategy for Improved Material Property Predictions. Journal of Chemical Theory and Computation, 20(15), 6756-6771. https://doi.org/10.1021/acs.jctc.4c00187
Vancouver
Du H, Hui J, Zhang L, Wang H. Rational Design of Deep Learning Networks Based on a Fusion Strategy for Improved Material Property Predictions. Journal of Chemical Theory and Computation. 2024;20(15):6756-6771. doi:10.1021/acs.jctc.4c00187
BibTeX
@article{Du2024,
author = {Du, Hongwei and Hui, Jian and Zhang, Lanting and Wang, Hong},
title = {Rational Design of Deep Learning Networks Based on a Fusion Strategy for Improved Material Property Predictions},
journal = {Journal of Chemical Theory and Computation},
year = {2024},
volume = {20},
number = {15},
pages = {6756--6771},
doi = {10.1021/acs.jctc.4c00187}
}
EUVIMED