Robust Field-level Likelihood-free Inference with Galaxies

Abstract

We train graph neural networks to perform field-level likelihood-free inference using galaxy catalogs from state-of-the-art hydrodynamic simulations of the CAMELS project. Our models are rotational, translational, and permutation invariant and do not impose any cut on scale. From galaxy catalogs that only contain $3$D positions and radial velocities of $\sim 1,000$ galaxies in tiny $(25h^{-1}\mathrm{Mpc}{)}^3$ volumes our models can infer the value of ${\Omega }_{\mathrm{m}}$ with approximately $12$ % precision. More importantly, by testing the models on galaxy catalogs from thousands of hydrodynamic simulations, each having a different efficiency of supernova and AGN feedback, run with five different codes and subgrid models - IllustrisTNG, SIMBA, Astrid, Magneticum, SWIFT-EAGLE -, we find that our models are robust to changes in astrophysics, subgrid physics, and subhalo/galaxy finder. Furthermore, we test our models on $1,024$ simulations that cover a vast region in parameter space - variations in $5$ cosmological and $23$ astrophysical parameters - finding that the model extrapolates really well. Our results indicate that the key to building a robust model is the use of both galaxy positions and velocities, suggesting that the network have likely learned an underlying physical relation that does not depend on galaxy formation and is valid on scales larger than $\sim 10h^{-1}\mathrm{kpc}$.

arXiv comment34 pages, 12 figures. For a video summarizing the results, see https://youtu.be/b59ep7cyPOs

References in corpus (70)

• The cosmological simulation code GADGET-2
• The EAGLE project: Simulating the evolution and assembly of galaxies and their environments
• Energy input from quasars regulates the growth and activity of black holes and their host galaxies
• Populating a cluster of galaxies - I. Results at z=0
• E pur si muove: Galiliean-invariant cosmological hydrodynamical simulations on a moving mesh
• Nine-Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Final Maps and Results
• Cosmological SPH simulations: A hybrid multi-phase model for star formation
• Modeling feedback from stars and black holes in galaxy mergers
• Simulating Galaxy Formation with the IllustrisTNG Model
• First results from the IllustrisTNG simulations: matter and galaxy clustering
• First results from the IllustrisTNG simulations: the stellar mass content of groups and clusters of galaxies
• First results from the IllustrisTNG simulations: the galaxy color bimodality
• The EAGLE simulations of galaxy formation: calibration of subgrid physics and model variations
• First results from the IllustrisTNG simulations: A tale of two elements -- chemical evolution of magnesium and europium
• Simulating galaxy formation with black hole driven thermal and kinetic feedback
• GIZMO: A New Class of Accurate, Mesh-Free Hydrodynamic Simulation Methods
• Simba: Cosmological Simulations with Black Hole Growth and Feedback
• First results from the IllustrisTNG simulations: radio haloes and magnetic fields
• Substructures in hydrodynamical cluster simulations
• Power Spectrum Analysis of Three-Dimensional Redshift Surveys
• Cosmological SPH simulations: The entropy equation
• Cosmology and fundamental physics with the Euclid satellite
• Extragalactic Science, Cosmology and Galactic Archaeology with the Subaru Prime Focus Spectrograph (PFS)
• The density and peculiar velocity fields of nearby galaxies
• Sample Variance Considerations for Cluster Surveys
• Cosmological simulations of black hole growth: AGN luminosities and downsizing
• The Arepo public code release
• Chemical enrichment of galaxy clusters from hydrodynamical simulations
• Turbulent gas motions in galaxy cluster simulations: The role of SPH viscosity
• The Quijote simulations
• The CAMELS project: Cosmology and Astrophysics with MachinE Learning Simulations
• The JWST Early Release Observations
• Putting the Precision in Precision Cosmology: How accurate should your data covariance matrix be?
• Super-Sample Covariance in Simulations
• Thermal Conduction in Simulated Galaxy Clusters
• On measuring the covariance matrix of the nonlinear power spectrum from simulations
• Constraining $M_{\nu }$ with the Bispectrum I: Breaking Parameter Degeneracies
• Origin and properties of dual and offset active galactic nuclei in a cosmological simulation at z=2
• Simulating the physical properties of dark matter and gas inside the cosmic web
• Fisher for complements: Extracting cosmology and neutrino mass from the counts-in-cells PDF
• Probing dark energy with cluster counts and cosmic shear power spectra: including the full covariance
• Nearest Neighbor distributions: new statistical measures for cosmological clustering
• X-ray mass proxies from hydrodynamic simulations of galaxy clusters (paper I)
• Hunting for Galaxies and Halos in simulations with VELOCIraptor
• Cosmicflows-4: The Catalog of ~10000 Tully-Fisher Distances
• The Euclid mission design
• Cosmological Forecasts for Combined and Next Generation Peculiar Velocity Surveys
• Constraining the astrophysics and cosmology from 21cm tomography using deep learning with the SKA
• The CAMELS Multifield Dataset: Learning the Universe's Fundamental Parameters with Artificial Intelligence
• The CAMELS project: public data release
• Massive data compression for parameter-dependent covariance matrices
• GalaxyNet: Connecting galaxies and dark matter haloes with deep neural networks and reinforcement learning in large volumes
• Joint analysis of anisotropic power spectrum, bispectrum and trispectrum: application to N-body simulations
• Machine Learning and Cosmological Simulations II: Hydrodynamical Simulations
• Introducing a new, robust galaxy finder algorithm for simulations
• Fast and Credible Likelihood-Free Cosmology with Truncated Marginal Neural Ratio Estimation
• SWIFT: Using task-based parallelism, fully asynchronous communication, and graph partition-based domain decomposition for strong scaling on more than 100,000 cores
• Fourier analysis of multi-tracer cosmological surveys
• A machine learning approach to mapping baryons onto dark matter haloes using the EAGLE and C-EAGLE simulations
• A Hybrid Deep Learning Approach to Cosmological Constraints From Galaxy Redshift Surveys
• Cosmology with one galaxy?
• Machine-assisted Semi-Simulation Model (MSSM): Estimating Galactic Baryonic Properties from their Dark Matter using a Machine Trained on Hydrodynamic Simulations
• $\text{Mangrove}$: Learning Galaxy Properties from Merger Trees
• The Cosmic Graph: Optimal Information Extraction from Large-Scale Structure using Catalogues
• Constraining the reionization history using deep learning from 21cm tomography with the Square Kilometre Array
• Finding universal relations in subhalo properties with artificial intelligence
• Inferring galaxy dark halo properties from visible matter with Machine Learning
• Multi-Epoch Machine Learning 1: Unravelling Nature vs Nurture for Galaxy Formation
• Velocity Correlations of Galaxy Clusters
• High-fidelity reproduction of central galaxy joint distributions with Neural Networks

Cited by in corpus (18)

• LtU-ILI: An All-in-One Framework for Implicit Inference in Astrophysics and Cosmology
• EFTofLSS meets simulation-based inference: ${\sigma }_8$ from biased tracers
• Data Compression and Inference in Cosmology with Self-Supervised Machine Learning
• Quijote-PNG: The Information Content of the Halo Mass Function
• Towards unveiling the large-scale nature of gravity with the wavelet scattering transform
• $\text{PineTree}$: A generative, fast, and differentiable halo model for wide-field galaxy surveys
• Learning the Universe: Cosmological and Astrophysical Parameter Inference with Galaxy Luminosity Functions and Colours
• Field-level simulation-based inference with galaxy catalogs: the impact of systematic effects
• Significance of void shape: Neutrino mass from Voronoi void halos?
• Trial by FIRE: Probing the dark matter density profile of dwarf galaxies with GraphNPE
• Inferring Cosmological Parameters on SDSS via Domain-Generalized Neural Networks and Lightcone Simulations
• Cosmology from Point Clouds with Dark Matter Halos from the Quijote Simulations
• CASCO: Cosmological and AStrophysical parameters from Cosmological simulations and Observations -- II. Constraining cosmology and astrophysical processes with early- and late-type galaxies
• Cosmology with Persistent Homology: Parameter Inference via Machine Learning
• Cosmology with Topological Deep Learning
• An Observationally Driven Multifield Approach for Probing the Circum-Galactic Medium with Convolutional Neural Networks
• Towards an optimal extraction of cosmological parameters from galaxy cluster surveys using convolutional neural networks
• Cosmological Analysis with Calibrated Neural Quantile Estimation and Approximate Simulators