Abstract

The 3D spatial distribution of galaxies encodes key cosmological information that can be used to probe the nature of dark energy and measure the sum of neutrino masses. The next generation of galaxy surveys, such as the Dark Energy Spectroscopic Instrument (DESI) and the Prime Focus Spectrograph (PFS), will observe 50 million galaxies over unprecedented cosmic volumes and produce the most precise measurements of galaxy clustering across 10 billion years of cosmic history. In my talk, I will present how we can leverage machine learning (ML) to go beyond current analyses and extract the full cosmological information of these galaxy surveys. In particular, I will present SimBIG, a framework for analyzing galaxy clustering using ML-based simulation-based inference. I will show the latest results from applying SimBIG to SDSS-III: BOSS observations and demonstrate that we can more than double the precision of current analyses. Lastly, I will present the status of the DESI and PFS surveys and how I will apply SimBIG to them to produce the leading constraints on dark energy and the sum of neutrino masses.