Event
CHBE Seminar: Dr. Debra Audus, NIST
Friday, March 28, 2025
11:00 a.m.
Room 2108 Chemical and Nuclear Engineering Building
Patricia Lorenzana
301-405-1935
plorenza@umd.edu
"Improving machine learning with polymer science."
Abstract: Machine learning as applied to polymer science has shown immense progress, mostly in areas where there are existing large datasets or where datasets can be generated quickly. However, there are numerous interesting problems where the dataset sizes are too small or the need to understand the science behind the machine learning prediction is essential. Here, we aim to tackle both problems by incorporating domain knowledge, in the form of polymer theories, into machine learning models. First, we consider a toy system of polymers in different solvent qualities and compare several methods for incorporating theory into machine learning using a simple, imperfect but easily interpretable theory. Second, we consider the phase behavior of polystyrene in cyclohexane using the foundational Flory-Huggins theory for guidance. We compare three theory informed machine learning methods along with a theory constrained machine learning method. Relative trade-offs and implications on explainability are explored. Finally, we also consider the problem of how to encode polymer architecture to produce the most accurate predictions for the cloud point of polymer solutions where the polymer can be linear, star or grafted to a nanoparticle.
Bio: Dr. Debra J. Audus is the project leader of the Polymer Analytics Project in the Materials Science and Engineering Division at NIST. She came to NIST in 2013 after receiving her PhD in Chemical Engineering at University of California, Santa Barbara and BS from Cornell University. Her research focuses on improving machine learning by embedding polymer science, explainable machine learning, polymer databases and using both theory and simulation to understand polymer physics. She is an active member of the American Physical Society, American Chemical Society and Materials Research Data Alliance, for which she organized the annual meeting in 2024. In 2023, she received the Communications Materials Outstanding Referee award.
