Journal Article · · Scientific Reports
DOI:https://doi.org/10.1038/s41598-023-36336-5· OSTI ID:2421183
Gupta, Vishu; Peltekian, Alec; Liao, Wei-keng; Choudhary, Alok; Agrawal, Ankit
Abstract Modern machine learning (ML) and deep learning (DL) techniques using high-dimensional data representations have helped accelerate the materials discovery process by efficiently detecting hidden patterns in existing datasets and linking input representations to output properties for a better understanding of the scientific phenomenon. While a deep neural network comprised of fully connected layers has been widely used for materials property prediction, simply creating a deeper model with a large number of layers often faces with vanishing gradient problem, causing a degradation in the performance, thereby limiting usage. In this paper, we study and propose architectural principles to address the question of improving the performance of model training and inference under fixed parametric constraints. Here, we present a general deep-learning framework based on branched residual learning (BRNet) with fully connected layers that can work with any numerical vector-based representation as input to build accurate models to predict materials properties. We perform model training for materials properties using numerical vectors representing different composition-based attributes of the respective materials and compare the performance of the proposed models against traditional ML and existing DL architectures. We find that the proposed models are significantly more accurate than the ML/DL models for all data sizes by using different composition-based attributes as input. Further, branched learning requires fewer parameters and results in faster model training due to better convergence during the training phase than existing neural networks, thereby efficiently building accurate models for predicting materials properties.
Cite
Export
Share
Save
Details
References (43)
Similar Records / Subjects
- Research Organization:
- Northwestern Univ., Evanston, IL (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC)
- DOE Contract Number:
- SC0019358; SC0021399
- OSTI ID:
- 2421183
- Journal Information:
- Scientific Reports, Vol. 13, Issue 1; ISSN 2045-2322
- Publisher:
- Nature Publishing Group
- Country of Publication:
- United States
- Language:
- English
References (43)
Similar Records
Improving deep learning model performance under parametric constraints for materials informatics applications
Journal Article · Mon Jun 05 00:00:00 EDT 2023 · Scientific Reports · OSTI ID:2421183
Gupta, Vishu; Peltekian, Alec; Liao, Wei-keng; +2 more
Simultaneously improving accuracy and computational cost under parametric constraints in materials property prediction tasks
Journal Article · Fri Feb 16 00:00:00 EST 2024 · Journal of Cheminformatics · OSTI ID:2421183
Gupta, Vishu; Li, Youjia; Peltekian, Alec; +4 more
Enabling deeper learning on big data for materials informatics applications
Journal Article · Fri Feb 19 00:00:00 EST 2021 · Scientific Reports · OSTI ID:2421183
Jha, Dipendra; Gupta, Vishu; Ward, Logan; +6 more
Related Subjects
Science & Technology - Other Topics