A machine learning approach to galaxy properties: joint redshift-stellar mass probability distributions with Random Forest

Abstract

We demonstrate that highly accurate joint redshift-stellar mass probability distribution functions (PDFs) can be obtained using the Random Forest (RF) machine learning (ML) algorithm, even with few photometric bands available. As an example, we use the Dark Energy Survey (DES), combined with the COSMOS2015 catalogue for redshifts and stellar masses. We build two ML models: one containing deep photometry in the $griz$ bands, and the second reflecting the photometric scatter present in the main DES survey, with carefully constructed representative training data in each case. We validate our joint PDFs for $10,699$ test galaxies by utilizing the copula probability integral transform and the Kendall distribution function, and their univariate counterparts to validate the marginals. Benchmarked against a basic set-up of the template-fitting code BAGPIPES, our ML-based method outperforms template fitting on all of our predefined performance metrics. In addition to accuracy, the RF is extremely fast, able to compute joint PDFs for a million galaxies in just under $6$ min with consumer computer hardware. Such speed enables PDFs to be derived in real time within analysis codes, solving potential storage issues. As part of this work we have developed GALPRO, a highly intuitive and efficient Python package to rapidly generate multivariate PDFs on-the-fly. GALPRO is documented and available for researchers to use in their cosmology and galaxy evolution studies.

18 pages, 8 figures, Accepted by MNRAS

References in corpus (77)

• emcee: The MCMC Hammer
• Stellar population synthesis at the resolution of 2003
• The Dust Content and Opacity of Actively Star-Forming Galaxies
• MultiNest: an efficient and robust Bayesian inference tool for cosmology and particle physics
• The 2.5 m Telescope of the Sloan Digital Sky Survey
• The Cosmic Evolution Survey (COSMOS) -- Overview
• EAZY: A Fast, Public Photometric Redshift Code
• Multimodal nested sampling: an efficient and robust alternative to MCMC methods for astronomical data analysis
• CIGALE: a python Code Investigating GALaxy Emission
• Bayesian photometric redshift estimation
• The COSMOS2015 Catalog: Exploring the 1<z<6 Universe with half a million galaxies
• The Dark Energy Camera
• The Dark Energy Survey: more than dark energy - an overview
• An ultra-deep near-infrared spectrum of a compact quiescent galaxy at z=2.2
• Measuring and Modelling the Redshift Evolution of Clustering: the Hubble Deep Field North
• Importance Nested Sampling and the MultiNest Algorithm
• Analysis of galaxy SEDs from far-UV to far-IR with CIGALE: Studying a SINGS test sample
• Modeling the Panchromatic Spectral Energy Distributions of Galaxies
• LSST Science Book, Version 2.0
• Euclid Definition Study Report
• The Connection between Galaxies and their Dark Matter Halos
• Inferring the star-formation histories of massive quiescent galaxies with BAGPIPES: Evidence for multiple quenching mechanisms
• Star Formation and Dust Attenuation Properties in Galaxies from a Statistical UV-to-FIR Analysis
• ANNz: estimating photometric redshifts using artificial neural networks
• A Modified Magnitude System that Produces Well-Behaved Magnitudes, Colors, and Errors Even for Low Signal-to-Noise Ratio Measurements
• Power Spectrum Tomography with Weak Lensing
• Modelling and interpreting spectral energy distributions of galaxies with BEAGLE
• Stellar Population Diagnostics of Elliptical Galaxy Formation
• Fitting the integrated Spectral Energy Distributions of Galaxies
• Photometric Redshifts for Hyper Suprime-Cam Subaru Strategic Program Data Release 1
• Slicing Through Multicolor Space: Galaxy Redshifts From Broadband Photometry
• First measurement of the Hubble constant from a dark standard siren using the Dark Energy Survey galaxies and the LIGO/Virgo binary-black-hole merger GW170814
• The Zurich Extragalactic Bayesian Redshift Analyzer (ZEBRA) and its first application: COSMOS
• TPZ : Photometric redshift PDFs and ancillary information by using prediction trees and random forests
• On the mass function of star clusters
• Dark Energy Survey Year 3 Results: Photometric Data Set for Cosmology
• Estimating photometric redshifts with artificial neural networks
• ANNz2 - photometric redshift and probability distribution function estimation using machine learning
• Photometric redshifts from SDSS images using a Convolutional Neural Network
• Photometric redshift estimation via deep learning
• Redshift distributions of galaxies in the DES Science Verification shear catalogue and implications for weak lensing
• Precision photometric redshift calibration for galaxy-galaxy weak lensing
• Machine Learning in Astronomy: a practical overview
• ArborZ: Photometric Redshifts Using Boosted Decision Trees
• Estimating Photometric Redshifts Using Support Vector Machines
• A statistical standard siren measurement of the Hubble constant from the LIGO/Virgo gravitational wave compact object merger GW190814 and Dark Energy Survey galaxies
• SOMz: photometric redshift PDFs with self organizing maps and random atlas
• Deconstructing the Galaxy Stellar Mass Function with UKIDSS and CANDELS: the Impact of Colour, Structure and Environment
• Euclid preparation: X. The Euclid photometric-redshift challenge
• The Cross-Correlation between Galaxies and Groups: Probing the Galaxy Distribution in and around Dark Matter Haloes
• No galaxy left behind: accurate measurements with the faintest objects in the Dark Energy Survey
• Unsupervised self-organised mapping: a versatile empirical tool for object selection, classification and redshift estimation in large surveys
• Evaluation of probabilistic photometric redshift estimation approaches for The Rubin Observatory Legacy Survey of Space and Time (LSST)
• Novel Methods for Predicting Photometric Redshifts from Broad Band Photometry using Virtual Sensors
• Using neural networks to estimate redshift distributions. An application to CFHTLenS
• A Sparse Gaussian Process Framework for Photometric Redshift Estimation
• Dark Energy Survey Year 3 Results: Measuring the Survey Transfer Function with Balrog
• The PAU Survey: Early demonstration of photometric redshift performance in the COSMOS field
• BayeSED: A General Approach to Fitting the Spectral Energy Distribution of Galaxies
• Robust Machine Learning Applied to Astronomical Datasets II: Quantifying Photometric Redshifts for Quasars Using Instance-Based Learning
• Star formation rates and stellar masses from machine learning
• Incorporating Photometric Redshift Probability Density Information into Real-Space Clustering Measurements
• Stellar mass as a galaxy cluster mass proxy: application to the Dark Energy Survey redMaPPer clusters
• GAz: A Genetic Algorithm for Photometric Redshift Estimation
• Photo-z Performance for Precision Cosmology
• Decoding spectral energy distributions of dust-obscured starburst-AGN
• A Comprehensive Bayesian Discrimination of the Simple Stellar Population Model, Star Formation History and Dust Attenuation Law in the Spectral Energy Distribution Modeling of Galaxies
• Can Self-Organizing Maps accurately predict photometric redshifts?
• Star Formation Rates for photometric samples of galaxies using machine learning methods
• Comparing Dark Energy Survey and HST-CLASH observations of the galaxy cluster RXC J2248.7-4431: implications for stellar mass versus dark matter
• Sacrificing information for the greater good: how to select photometric bands for optimal accuracy
• How to measure metallicity from five-band photometry with supervised machine learning algorithms
• MAGPHYS: a publicly available tool to interpret observed galaxy SEDs
• Uncertain Photometric Redshifts
• Evolution of Galaxy Luminosity and Stellar-Mass Functions since $z=1$ with the Dark Energy Survey Science Verification Data
• Dark Energy Survey Year 3 Results: Deep Field Optical + Near-Infrared Images and Catalogue
• Stellar Masses of High-Redshift Galaxies

Cited by in corpus (22)

• A standard siren measurement of the Hubble constant using gravitational wave events from the first three LIGO/Virgo observing runs and the DESI Legacy Survey
• Dark Energy Survey Year 3 Results: Measuring the Survey Transfer Function with Balrog
• Photometric redshift-aided classification using ensemble learning
• Implicit Likelihood Inference of Reionization Parameters from the 21 cm Power Spectrum
• Photometric redshifts with machine learning, lights and shadows on a complex data science use case
• Augmenting astrophysical scaling relations with machine learning: application to reducing the Sunyaev-Zeldovich flux-mass scatter
• What to expect from dynamical modelling of cluster haloes II. Investigating dynamical state indicators with Random Forest
• Euclid preparation: XXII. Selection of Quiescent Galaxies from Mock Photometry using Machine Learning
• Non-parametric reconstruction of cosmological observables using Gaussian Processes Regression
• A Machine Learning Approach to Predict Missing Flux Densities in Multi-band Galaxy Surveys
• Estimation of the masses in the Local Group by Gradient Boosted Decision Trees
• Improving machine learning-derived photometric redshifts and physical property estimates using unlabelled observations
• Euclid preparation. LI. Forecasting the recovery of galaxy physical properties and their relations with template-fitting and machine-learning methods
• Identifying type II quasars at intermediate redshift with few-shot learning photometric classification
• A statistical study of lopsided galaxies using random forest
• Dark from light (DfL): Inferring halo properties from luminous tracers with machine learning trained on cosmological simulations. I. Method, proof of concept & preliminary testing
• Euclid preparation. Estimating galaxy physical properties using CatBoost chained regressors with attention
• Interpreting deep learning-based stellar mass estimation via causal analysis and mutual information decomposition
• Constraining Galaxy-Halo Connection Using Machine Learning
• Predicting halo occupation and galaxy assembly bias with machine learning
• Non-Sequential Neural Network for Simultaneous, Consistent Classification and Photometric Redshifts of OTELO Galaxies
• A Value-added Physical Properties Catalog for Low-redshift Galaxies from DESI Legacy Imaging Surveys DR10