Title: Using machine learning to look for dark matter in the Galactic Center, the Milky Way halo, and other galaxies
Abstract: Advancements in machine learning have enabled new ways of doing inference on forward models defined through complex, high-dimensional simulations. After briefly motivating their use in the cosmological context, I will present applications of these simulation-based inference methods to three separate astrophysical systems with the goal of looking for signatures of dark matter. First, I will describe how they can be used to combine information from thousands of strong gravitational lensing systems in a principled way in order to extract the population properties of dark matter. Then, I will quantify their sensitivity to the collective imprint of dark matter subhalos in our own Galaxy on the measured motions of background luminous celestial objects. Finally, I will present an application to the Galactic Center gamma-ray excess, where the fact that our method can extract more information from the gamma-ray dataset than is possible with traditional techniques can be used to more robustly characterize the nature of the excess and constrain its dark matter properties.
For help, please contact Webmaster.