CASBI: Chemical Abundance Simulation Based Inference

Galaxies evolve hierarchically through merging with lower-mass systems, and the ``remnants" of these events are a key indicator of their past assembly history. These ``remnants" are star systems that, during their lifetime, were captured by the gravitational potential of a larger galaxy, and ended up in the star halo of their host. Accurately measuring the properties of the accreted galaxies and hence unraveling the Milky Way (MW) formation history is a challenging task. In this work I present CASBI (Chemical Abundance Simulation Based Inferece), a novel inference pipeline for Galactic Archaeology based on Simulation Based Inference methods. CASBI leverages on the fact that there is a well defined mass-metallicity relation for galaxies. CASBI performs inference of key galaxy properties based on multidimensional chemical abundances of stars in the stellar halo. Hence, I recast the task of unraveling the merger history of the MW into a SBI problem to recover the properties (e.g. total stellar mass and infall time) of the halo building blocks using the multidimensional chemical abundances of stars in the stellar halo as observable. I highlight CASBI's potential by inferring posteriors for the stellar masses of completely phase mixed dwarf galaxies solely from the 2d-distributions of stellar abundance in the iron vs. oxygen plane and find accurate and precise inference results.