Campus, PAB 214
High-resolution hydrodynamic zoom-in simulations of Milky Way-mass halos offer exquisite resolution and provide insights into the small-scale challenges associated with ΛCDM cosmology. However, these simulations are computationally expensive, and the relationship between different sub-grid baryonic physics prescriptions and their impact on subhalo and satellite galaxy populations is not well understood. In this talk, we present a machine learning algorithm trained on hydrodynamic simulations that efficiently predicts surviving subhalo populations from dark matter only simulations. We discuss applications of this technique to predict satellite luminosity functions for Milky Way analogs and related work on modeling the dwarf satellite population of the Milky Way.