Speaker
Mr
Driss Mestiri
Description
I present a simulation-based inference framework that characterizes dark matter microphysics through GD-1 stellar stream perturbations. By training advanced neural networks on millions of simulated streams, the pipeline can return calibrated parameter posteriors and model-class probabilities for cold, self-interacting and ultra-light dark matter, as well as baryonic perturbers down to 3×10⁶ M⊙. Ultimately delivering quantitative constraints on the Milky Way substructure spectrum, and motivating refined modeling and future multi-probe tests of dark matter physics.
| What area of study best describes your talk? | Analysis |
|---|
Author
Co-author
Mr
Pranav Kulkarni
(Stanford Artificial Intelligence Laboritory)