Learning cosmology and clustering with cosmic graphs

TL;DR

This paper demonstrates that Graph Neural Networks can effectively analyze galaxy distributions from cosmological simulations, accurately estimating cosmological parameters and power spectra while accounting for astrophysical uncertainties.

Contribution

It introduces GNN architectures tailored for galaxy data, achieving high-precision cosmological inference directly from galaxy positions and properties.

Findings

GNNs can compute the power spectrum with a few percent accuracy.

Likelihood-free inference of $ m \Omega_m$ with 12-13% accuracy from galaxy positions.

Inclusion of galaxy properties improves inference accuracy to 4-8%.

Abstract

We train deep learning models on thousands of galaxy catalogues from the state-of-the-art hydrodynamic simulations of the CAMELS project to perform regression and inference. We employ Graph Neural Networks (GNNs), architectures designed to work with irregular and sparse data, like the distribution of galaxies in the Universe. We first show that GNNs can learn to compute the power spectrum of galaxy catalogues with a few percent accuracy. We then train GNNs to perform likelihood-free inference at the galaxy-field level. Our models are able to infer the value of $\Omega_{\rm m}$ with a $\sim12\%-13\%$ accuracy just from the positions of $\sim1000$ galaxies in a volume of $(25~h^{-1}{\rm Mpc})^3$ at $z=0$ while accounting for astrophysical uncertainties as modelled in CAMELS. Incorporating information from galaxy properties, such as stellar mass, stellar metallicity, and stellar radius, increases the accuracy to $4\%-8\%$. Our models are built to be translational and rotational invariant, and they can extract information from any scale larger than the minimum distance between two galaxies. However, our models are not completely robust: testing on simulations run with a different subgrid physics than the ones used for training does not yield as accurate results.