Speaker
Description
The 3D distribution of galaxies encodes key cosmological information that can probe the growth and expansion history of the Universe. In my talk, I will present how we can leverage simulations and machine learning to go beyond current analyses and extract the full cosmological information of the next-generation galaxy surveys. In particular, I will present SimBIG, a forward modeling framework for analyzing galaxy clustering using simulation-based inference based on normalizing flows. I will show the latest results from applying SimBIG to BOSS observations to analyze the bispectrum, wavelet scattering transform, and a field-level summary based on convolutional neural networks— all down to small, non-linear, scales. By robustly extracting additional cosmological information, we constrain $\Lambda$CDM parameters, $\Omega_b$, $h$, $n_s$, $\Omega_m$, and $\sigma_8$, that are 2.4, 1.5, 1.7, 1.2, and 2.7$\times$ tighter than standard power spectrum analyses. With this increased precision, we derive constraints on the Hubble constant, $H_0$, and $S_8 = \sigma_8 \sqrt{\Omega_m/0.3}$ that are competitive with other cosmological probes and inform cosmic tensions, even with a sample that only spans 10\% of the full BOSS volume. Lastly, I will discuss how SimBIG can be extended to upcoming spectroscopic galaxy surveys (DESI, PFS, Euclid) to produce leading $H_0$ and $S_8$ constraints.
