I'm an assistant professor at Princeton University's Department of Astrophysical Sciences and the Center for Statistics and Machine Learning. I lead the Princeton Astro Data Lab, where we solve problems that hold back astronomy, and the Dynamical Learning Lab, where we model complex evolving systems, from nature to neural networks.

I design methods to push the limits of what we can learn about physical systems from data that are noisy, incomplete, censored, glitchy, contaminated, and often all of the above. How? By assuming that physics works, even though any theoretical model is incomplete. My group creates novel techniques that embed physical principles in deep learning architectures. We develop methods for signal separation, data fusion, fast inference, and outlier detection for large astronomical surveys. I'm funded by NSF and NASA to develop joint analysis methods for ground-based and space-based images (between Rubin, Euclid, and Roman); by the Keck Foundation to discover Earth-like exoplanets; and by Schmidt Sciences to optimize the utilization of spectra for PFS.

My research on physics-ML hybrids also applies to remote and environmental sensing. For the NSF-funded HydroGEN collaboration, I lead the method development of a physics-based machine learning platform for hydrologic scenario generation to predict droughts, wildfire conditions, and the availability of drinking water for the entire continental United States.

For Students
If you're already at Princeton: I usually have projects for undergraduate or graduate students, so come talk to me. As PhD applicant, please directly apply to the department. After you have been accepted, I'm happy to talk about projects.