Research Overview

The astronomy community is preparing to collect stunningly detailed multi-wavelength maps of the sky, and new data analysis methods are needed to take advantage of such rich data. To address this, powerful new machine learning techniques are being developed. But moving from traditional data analysis to modern machine learning is a shift in our scientific approach, and the shift requires us to ask an important question: Can we trust the black box? My research meets this need by creating the machine learning methods for interpreting huge volumes of cosmological data. My research also builds trust in those methodologies: rather than using machine learning as a black box, I develop novel interpretation schemes so that the techniques can be understood, trusted, and applied with confidence.

Selected Publications

See all publications: Google Scholar, arXiv, ADS, ORCiD.