I presented my abstract On Differentiable Correlation Functions at the American Astronomical Society Winter Session Meeting in Baltimore, MD. My work is important because it enables gradient-based optimization, enabling cosmologists to relate correlations to astrophysical model parameters. In my current extensions to this work, I have shown that it enables a new class of uncertainty quantification and alternatively adds functionality to other machine learning approaches.
Those who attended my talk seemed to take great interest, asking if my algorithm had applications outside of correlation functions to which I answered yes! My favorite session that I attended was about the Habitable Worlds Observatory and the continued search for extraterrestrial life. My prior work on astronomical imaging was done in the context of weak lensing but can also be applied to exoplanet literature, which means it’s more widely applicable and impactful than I initially thought!
