Hunting for dark matter subhalos in the Milky Way with machine learning techniques

Rapid advances in the field of Machine Learning (ML) have provided
us with new tools for reducing and analyzing vast datasets. In my talk I
will discuss possible applications of ML techniques to learn more about the
nature of Dark Matter (DM) from recent and near future extensive
astronomical observations. In particular, I will focus on the possibility of
detecting dark galactic subhaloes (i.e. small DM subhaloes which host
negligible amount of stars and gas) through their gravitational impact on
the Milky Way's stellar field which is currently being mapped by the Gaia
satellite with unprecedented precision. Detection of these small DM
subhaloes would provide strong evidence in favor of the cold DM paradigm
while the opposite is true in case of their absence. I will also briefly
discuss the possibility to explore the DM content of luminous galaxies using
ML algorithms trained on modern hydrodynamical simulations.