ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Machine Learning Constrained with Dimensional Analysis and Scaling Laws: Simple, Transferable, and Interpretable Models of Materials from Small Datasets

Kumar, Narendra and Rajagopalan, Padmini and Pankajakshan, Praveen and Bhattacharyya, Arnab and Sanyal, Suchismita and Balachandran, Janakiraman and Waghmare, Umesh V (2019) Machine Learning Constrained with Dimensional Analysis and Scaling Laws: Simple, Transferable, and Interpretable Models of Materials from Small Datasets. In: CHEMISTRY OF MATERIALS, 31 (2). pp. 314-321.

[img] PDF
Che_Mat_31-2_314_2019.pdf - Published Version
Restricted to Registered users only

Download (1MB) | Request a copy
Official URL: https://doi.org/10.1021/acs.chemmater.8b02837

Abstract

Machine learning (ML) from materials data-bases can accelerate the design and discovery of new materials through the development of accurate, computationally inexpensive models to predict materials properties. These models in turn enable rapid screening of large materials search space. However, materials datasets describing functional properties are typically small, which creates challenges pertaining to interpretability and transferability when exploring them with conventional ML approaches. Further, correlations within the dataset can lead to instability (nonunique functional models relating inputs to outputs) and overfitting. In this work, we address these issues by developing a new approach, in which ML with the Bootstrapped projected gradient descent algorithm is constrained with Buckingham Pi theorem-based dimensional analysis and scaling laws of relationships between different input descriptors (properties). This constrained learning model enables us to learn from small data and develop predictive models that are accurate, computationally inexpensive, and physically interpretable. We demonstrate this approach by developing a simple model to predict the intrinsic dielectric breakdown field based on an available dataset of 82 compounds. Our approach is generic in nature and is expected to work effectively with other sparse materials datasets.

Item Type: Journal Article
Additional Information: Copyright of this article belongs to AMER CHEMICAL SOC
Department/Centre: Division of Electrical Sciences > Computer Science & Automation
Depositing User: Id for Latest eprints
Date Deposited: 14 Feb 2019 05:52
Last Modified: 14 Feb 2019 05:52
URI: http://eprints.iisc.ac.in/id/eprint/61701

Actions (login required)

View Item View Item