Welcome to my research page! My name is Gage and I’m a postdoc in the Princeton University Physics and ORFE departments, working with Boris Hanin and Dan Marlow. I’m interested in developing principles for machine learning initialization / architecture selection and applying them to physics tasks. Specifically, I am interested in generating principles for building robust and easily-trainable equivariant models and applying them to downstream scientific applications.

I recently defended my PhD thesis describing a search for a pseudoscalar Higgs boson using Run 2 data collected at the CMS detector. Along the way, I also applied graph neural networks to charged particle tracking (see some recent results in collaboration with Kilian Lieret here) and developed principles for initialization and architecture selection in GNNs with Boris; you can view a preprint of our recent article here: arxiv:2306.11668.

Outside of research, I’ve organized a Machine Learning in Physics seminar for the 2023-2024 academic year along with Boris, Dan, Mariangela Lisanti, and Peter Elmer. The focus is on interdisciplinary work, both physics for ML and ML for physics. We’ve prepared a fantastic lineup of speakers, check out our seminar webpage for more details.

I’m also teaching two sections of Physics 103: General Physics I this semester; this is an introductory mechanics course designed for a broad range of STEM majors.

If any of this sounds interesting, I’d love to hear from you! Please feel free to contact me at jdezoort@princeton.edu.