Professor Yanlai Chen (PI) and Chancellor Professor Sigal Gottlieb (Co-PI) have been awarded $296,555 for the project, "Reduced Basis Enhancements of Neural Networks and Their Application to Quantum Materials Simulation," by the National Science Foundation (NSF).
The need to understand the configuration-to-performance map of a system efficiently and accurately under various configurations is ubiquitous, yet challenging, due to the prohibitively high computational cost. This project aims to closely integrate two techniques to tackle this challenge by building analysis-driven computational emulators for these systems. The first technique is the more traditional and mathematically rigorous reduced basis method, and the other is the more nascent deep neural networks. The resulting emulators learn the system behavior reliably and are expected to perform better, than the current approaches, on data unseen during training.
As an application of the developed methodology, this project aims to provide a systematic and rigorous study of parameterized 2D materials simulation, including the recently discovered magical angle twisted bilayer graphene.
"I am grateful to the NSF for recognizing our vision, the great research our team of faculty and students have been conducting at UMass Dartmouth, and its further potential," said Chen. "The outcomes of this project are expected to benefit the greater scientific community that utilizes supervised machine learning for parameterized models. The project involves development of graduate coursework for the Engineering and Applied Science program and the training of undergraduate and graduate students through involvement in the research."
This is Chen’s fifth NSF grant and third along the topic of research on Reduced Basis Method.Filed under: College of Arts and Sciences, Departments : Directory Mathematics Dept