Joint Astrophysics Colloquium

Digging into the Large Scale Structure: From the galaxies to the cosmic web

Shirley Ho

Carnegie Mellon University

Galaxy spectroscopic surveys provide the means to map out this cosmic large-scale structure in three dimensions, furnishing a cornerstone of observational cosmology. The information is given in the form of galaxy locations, and is typically condensed into a single function of scale, such as the galaxy correlation function or power-spectrum. However, galaxy correlation functions are not the only information those surveys provide. One of the most striking features of N-body simulations is the network of filaments into which dark matter particles arrange themselves. We however traditionally only use the information contained in the positions of the galaxies, and in some occasions, we look at other cosmic structures of the Universe such as voids.

In this seminar, I explore the information beyond the galaxy positions in large sky surveys combining novel ideas with recent techniques in statistical methods and machine learning algorithms. In particular, we will investigate the following two topics: the "cosmic web" that are mostly ignored in any large scale structure analyses in the Universe and how it affects the surrounding galaxies; and "likelihood free analysis" with machine learning.