| Current astrophysics is characterized by the confluence of three separate developments: new instruments and large surveys providing more observational data than ever; massive simulations probing ever more complex phenomena; and rapid advances in machine learning establishing entirely new ways of dealing with and interpreting data. None of these three pillars can stand on their own to make meaningful progress in astronomy and the physical sciences. I will present three areas of research that combine accurate statistical modeling, deep neural networks, and numerical simulations: joint-survey processing of LSST, Euclid, and Roman data; simulation-based inference in cosmology and galaxy evolution; and science-driven design of new surveys and observing programs. |