Psi-GAN: A power-spectrum-informed generative adversarial network for the emulation of large-scale structure maps across cosmologies and redshifts

Abstract

Simulations of the dark matter distribution throughout the Universe are essential in order to analyse data from cosmological surveys. $N$-body simulations are computationally expensive, and many cheaper alternatives (such as lognormal random fields) fail to reproduce accurate statistics of the smaller, non-linear scales. In this work, we present \textsc{Psi-GAN} (\textbf{P}ower-\textbf{s}pectrum-\textbf{i}nformed \textbf{G}enerative \textbf{A}dversarial \textbf{N}etwork), a machine learning model which takes a two-dimensional lognormal dark matter density field and transforms it into a more realistic field. We construct \textsc{Psi-GAN} so that it is continuously conditional, and can therefore generate realistic realisations of the dark matter density field across a range of cosmologies and redshifts in $z\in [0,3]$. We train \textsc{Psi-GAN} as a generative adversarial network on $2000$ simulation boxes from the Quijote simulation suite. We use a novel critic architecture that utilises the power spectrum as the basis for discrimination between real and generated samples. \textsc{Psi-GAN} shows agreement with $N$-body simulations over a range of redshifts and cosmologies, consistently outperforming the lognormal approximation on all tests of non-linear structure, such as being able to reproduce both the power spectrum up to wavenumbers of $1h{\mathrm{Mpc}}^{-1}$, and the bispectra of target $N$-body simulations to within $\sim 5$ per cent. Our improved ability to model non-linear structure should allow more robust constraints on cosmological parameters when used in techniques such as simulation-based inference.

20 pages, 11 figures, 3 tables, 1 appendix. Accepted for publication by Monthly Notices of the Royal Astronomical Society

References in corpus (43)

• The cosmological simulation code GADGET-2
• Simulating the joint evolution of quasars, galaxies and their large-scale distribution
• Gaussian Error Linear Units (GELUs)
• The Cosmic Linear Anisotropy Solving System (CLASS) II: Approximation schemes
• A History of Dark Matter
• Resolving Cosmic Structure Formation with the Millennium-II Simulation
• Enzo: An Adaptive Mesh Refinement Code for Astrophysics
• A New Era in the Quest for Dark Matter
• Simulating cosmic structure formation with the GADGET-4 code
• The Quijote simulations
• The CAMELS project: Cosmology and Astrophysics with MachinE Learning Simulations
• Cosmology and the Bispectrum
• nbodykit: an open-source, massively parallel toolkit for large-scale structure
• Accurate Estimators of Correlation Functions in Fourier Space
• Learning to Predict the Cosmological Structure Formation
• Improving lognormal models for cosmological fields
• Cosmology constraints from shear peak statistics in Dark Energy Survey Science Verification data
• KiDS-450: Cosmological Constraints from Weak Lensing Peak Statistics - II: Inference from Shear Peaks using N-body Simulations
• CosmoGAN: creating high-fidelity weak lensing convergence maps using Generative Adversarial Networks
• KiDS-450: Cosmological Constraints from Weak Lensing Peak Statistics-I: Inference from Analytical Prediction of High Signal-to-Noise Ratio Convergence Peaks
• Fast cosmic web simulations with generative adversarial networks
• Fourier analysis of luminosity-dependent galaxy clustering
• Cosmic Shear Cosmology Beyond 2-Point Statistics: A Combined Peak Count and Correlation Function Analysis of DES-Y1
• Testing the lognormality of the galaxy and weak lensing convergence distributions from Dark Energy Survey maps
• A new model to predict weak-lensing peak counts I. Comparison with $N$-body Simulations
• A new model to predict weak-lensing peak counts II. Parameter constraint strategies
• Constraining Neutrino Mass with the Tomographic Weak Lensing Bispectrum
• Optimal capture of non-Gaussianity in weak lensing surveys: power spectrum, bispectrum and halo counts
• Field Level Neural Network Emulator for Cosmological N-body Simulations
• Cosmological constraints from the capture of non-Gaussianity in Weak Lensing data
• The bispectrum of polarized galactic foregrounds
• Universal Behavior of Phase Correlations in Non-linear Gravitational Clustering
• A new model to predict weak-lensing peak counts III. Filtering technique comparisons
• Phase Correlations in Non-Gaussian Fields
• Nonlinear 3D Cosmic Web Simulation with Heavy-Tailed Generative Adversarial Networks
• GLASS: Generator for Large Scale Structure
• Explaining deep learning of galaxy morphology with saliency mapping
• Statistics of Fourier Modes in Non-Gaussian Fields
• Fast and Accurate Non-Linear Predictions of Universes with Deep Learning
• Fast and realistic large-scale structure from machine-learning-augmented random field simulations
• Interpreting automatic AGN classifiers with saliency maps
• Scalable and Equivariant Spherical CNNs by Discrete-Continuous (DISCO) Convolutions
• GANSky -- fast curved sky weak lensing simulations using Generative Adversarial Networks

Cited by in corpus (1)

• Conditional variational autoencoders for cosmological model discrimination and anomaly detection in cosmic microwave background power spectra