Machine Learning Framework for Polymer Discovery
Document Type
Article
Publication Date
1-1-2022
Abstract
With rapid advances in computational power, machine learning (ML) has boomed as an effective tool to discover new materials. In the field of polymers, ML has also found its applications in finding new materials with desired performance in a lot of studies. In this article, we review a brief history for ML development and summarize a main framework for ML assisted polymer discovery. Next, we elaborate four key steps in the framework, i.e., database establishment, fingerprinting, model establishment, and new polymer molecule generation, and then review some commonly leveraged methodologies in these steps. Finally, the main challenges in ML assisted polymer discovery are discussed and some outlooks are discussed.
Publication Source (Journal or Book title)
Encyclopedia of Materials: Plastics and Polymers
First Page
267
Last Page
279
Recommended Citation
Yan, C., & Li, G. (2022). Machine Learning Framework for Polymer Discovery. Encyclopedia of Materials: Plastics and Polymers, 1-4, 267-279. https://doi.org/10.1016/B978-0-12-820352-1.00206-6
