Painting galaxies into dark matter halos using machine learning

Abstract

We develop a machine learning (ML) framework to populate large dark matter-only simulations with baryonic galaxies. Our ML framework takes input halo properties including halo mass, environment, spin, and recent growth history, and outputs central galaxy and halo baryonic properties including stellar mass ($M_{\ast }$), star formation rate (SFR), metallicity ($Z$), neutral ($\mathrm{HI}$) and molecular (${\mathrm{H}}_2$) hydrogen mass. We apply this to the MUFASA cosmological hydrodynamic simulation, and show that it recovers the mean trends of output quantities with halo mass highly accurately, including following the sharp drop in SFR and gas in quenched massive galaxies. However, the scatter around the mean relations is under-predicted. Examining galaxies individually, at $z=0$ the stellar mass and metallicity are accurately recovered ($\sigma \lesssim 0.2$~dex), but SFR and $\mathrm{HI}$ show larger scatter ($\sigma \gtrsim 0.3$~dex); these values improve somewhat at $z=1,2$. Remarkably, ML quantitatively recovers second parameter trends in galaxy properties, e.g. that galaxies with higher gas content and lower metallicity have higher SFR at a given $M_{\ast }$. Testing various ML algorithms, we find that none perform significantly better than the others, nor does ensembling improve performance, likely because none of the algorithms reproduce the large observed scatter around the mean properties. For the random forest algorithm, we find that halo mass and nearby ($\sim 200$~kpc) environment are the most important predictive variables followed by growth history, while halo spin and $\sim$Mpc scale environment are not important. Finally we study the impact of additionally inputting key baryonic properties $M_{\ast }$, SFR and $Z$, as would be available e.g. from an equilibrium model, and show that particularly providing the SFR enables $\mathrm{HI}$ to be recovered substantially more accurately.

15 pages, 10 figures, 1 table, accepted version from MNRAS

References in corpus (8)

• The EAGLE project: Simulating the evolution and assembly of galaxies and their environments
• Introducing the Illustris Project: Simulating the coevolution of dark and visible matter in the Universe
• Introducing the Illustris Project: the evolution of galaxy populations across cosmic time
• Dark Matter Substructure in Numerical Simulations: A Tale of Discreteness Noise, Runaway Instabilities, and Artificial Disruption
• MUFASA: Galaxy star formation, gas, and metal properties across cosmic time
• A Critical Look at the Mass-Metallicity-SFR Relation in the Local Universe. I. An Improved Analysis Framework and Confounding Systematics
• The Outer Halos of Very Massive Galaxies: BCGs and their DSC in the Magneticum Simulations
• Mufasa:The strength and evolution of galaxy conformity in various tracers

Cited by in corpus (50)

• The Dust-to-Gas and Dust-to-Metals Ratio in Galaxies from z=0-6
• Surveying the reach and maturity of machine learning and artificial intelligence in astronomy
• Mergers, Starbursts, and Quenching in the Simba Simulation
• AI-assisted super-resolution cosmological simulations
• Cosmological constraints from noisy convergence maps through deep learning
• Distinguishing standard and modified gravity cosmologies with machine learning
• Machine learning cosmological structure formation
• GalaxyNet: Connecting galaxies and dark matter haloes with deep neural networks and reinforcement learning in large volumes
• A machine learning approach to mapping baryons onto dark matter haloes using the EAGLE and C-EAGLE simulations
• 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
• Predicting dark matter halo formation in N-body simulations with deep regression networks
• Machine Learning for Observational Cosmology
• Learning to Concentrate: Multi-tracer Forecasts on Local Primordial Non-Gaussianity with Machine-Learned Bias
• The scatter in the galaxy-halo connection: a machine learning analysis
• Augmenting astrophysical scaling relations with machine learning: application to reducing the Sunyaev-Zeldovich flux-mass scatter
• From EMBER to FIRE: predicting high resolution baryon fields from dark matter simulations with Deep Learning
• Predicting the Neutral Hydrogen Content of Galaxies From Optical Data Using Machine Learning
• Generating Synthetic Cosmological Data with GalSampler
• Multiwavelength cluster mass estimates and machine learning
• Multi-Epoch Machine Learning 1: Unravelling Nature vs Nurture for Galaxy Formation
• Hybrid analytic and machine-learned baryonic property insertion into galactic dark matter haloes
• Large-scale structures in the $\Lambda$CDM Universe: network analysis and machine learning
• Teaching neural networks to generate Fast Sunyaev Zel'dovich Maps
• Modelling the galaxy-halo connection with semi-recurrent neural networks
• How the Galaxy-Halo Connection Depends on Large-Scale Environment
• Mock halo catalogs: assigning unresolved halo properties using correlations with local halo environment
• SHAMe-SF: Predicting the clustering of star-forming galaxies with an enhanced abundance matching model
• High-fidelity reproduction of central galaxy joint distributions with Neural Networks
• Not Hydro: Using Neural Networks to estimate galaxy properties on a Dark-Matter-Only simulation
• Classification algorithms applied to structure formation simulations
• MAHGIC: A Model Adapter for the Halo-Galaxy Inter-Connection
• Machine learning prediction for mean motion resonance behaviour -- The planar case
• Approximations to galaxy star formation rate histories: properties and uses of two examples
• A search for dark matter among Fermi-LAT unidentified sources with systematic features in Machine Learning
• Machine Learning the Fates of Dark Matter Subhalos: A Fuzzy Crystal Ball
• Multi-Epoch Machine Learning 2: Identifying physical drivers of galaxy properties in simulations
• Baryon Pasting the Uchuu Lightcone Simulation
• A proof-of-concept neural network for inferring parameters of a black hole from partial interferometric images of its shadow
• On the unique evolutionary mechanisms of massive quiescent galaxies in the epoch of reionisation
• A sparse regression approach for populating dark matter halos and subhalos with galaxies
• Large-step neural network for learning the symplectic evolution from partitioned data
• Merger Tree-based Galaxy Matching: A Comparative Study Across Different Resolutions
• Predicting halo occupation and galaxy assembly bias with machine learning
• Insights into the dependence of galaxy properties on the environment with explainable machine learning models
• JERALD: high-fidelity dark matter, stellar mass and neutral hydrogen maps from fast N-body simulations
• Constraining Galaxy-Halo Connection Using Machine Learning
• Populating Galaxies Into Halos Via Machine Learning on the Simba Simulation
• Predicting large scale cosmological structure evolution with generative adversarial network-based autoencoders
• Mimicking the halo-galaxy connection using machine learning