Modeling Nuclei via Machine Learning Networks
Join us in modeling a fundamental property of nuclei via the exciting new realm of Machine Learning (ML) networks! We model the nuclear mass via a novel Machine Learning algorithm, with the aim of finding underlying physical correlations in exotic, inaccessible nuclei.
Our algorithm, the Physically Interpretable Machine Learning (PIML) model, is a Python-based mixture density network model that determines the nuclear mass via a mapping of probability distributions. It is constantly being improved, updated and tinkered with, and we're looking for students to assist in its development and implementation.
Project scope is flexible, and ranges from front-end code usage to back-end algorithmic developments. While prior coding knowledge is useful, it is by no means required; regardless of your previous coding experience, we'll have a project ready to tailor to you!
If you're interested in our efforts in ML within nuclear physics, please join us!
Semester project opportunities are primarily computational, and can be flexed to fit what you might be looking for. Hours and commitment time are also flexible; we will work with you to create something optimized to what you're looking for!
We have a wealth of ideas and projects ready for motivated students; opportunities to lead a project push are readily available, including semi-independent investigation and potential publication
Our group consists of multiple faculty members and students who are all ready to work with you; you'll be joining a lively group of research peers!