PREDICTING COSMIC VOID PROFILES WITH DEEP LEARNING

Responsabile: Carmelita Carbone

Descrizione: Cosmic voids are the largest (non-virialized) structures in the Universe. The volume distribution of cosmic voids is the result of the hierarchical growth of initial density perturbations through accretion and mergers. In this thesis project we aim at developing a machine-learning-based model for the total density profiles of cosmic voids detected in the DEMNUni simulations. The model will provide predictions for the density profiles in different cosmological models (dynamical dark energy, massive neutrinos), including radial ranges and the outer region profile. The idea is to interpret them in terms of the formation history of cosmic voids. The new profile fit could be used in observational cosmology studies which aim to constrain cosmology from large-scale structures.

Durata: 6/9 mesi

Laurea: Magistrale

Prerequisiti: Esame di Cosmologia e nozioni di programmazione