Classifying Stars, Galaxies and AGN in CLAUDS+HSC-SSP Using Gradient Boosted Decision Trees

Abstract

Classifying catalog objects as stars, galaxies, or AGN is a crucial part of any statistical study of galaxies. We describe our pipeline for binary (star/galaxy) and multiclass (star/galaxy/Type I AGN/Type II AGN) classification developed for the very deep CLAUDS+HSC-SSP $u^{\ast }grizy$ dataset. Our method uses the XGBoost implementation of Gradient Boosted Trees (GBT) to train ensembles of models which take photometry, colours, maximum surface brightnesses, and effective radii from all available bands as input, and output the probability that an object belongs to each of the classes under consideration. At $i_{AB}<25$ our binary star/galaxy model has AUC=0.9974 and at the threshold that maximizes our sample's weighted F1 score, selects a sample of galaxies with 99.7% purity and 99.8% completeness. We test the model's ability to generalize to objects fainter than those seen during training and find that extrapolation of ~1-2 magnitudes is reasonable for most applications provided that the galaxies in the training sample are representative of the range of redshifts and colours of the galaxies in the target sample. We also perform an exploratory analysis of the method's ability to identify AGN using a small x-ray selected sample and find that it holds promise for classifying type I AGN, although it performs less well for type II AGN. Our results demonstrate that GBTs provide a flexible, robust and efficient method for performing classification of catalog objects in large astronomical imaging surveys.

arXiv commentAccepted for publication in MNRAS

References in corpus (43)

• Scikit-learn: Machine Learning in Python
• XGBoost: A Scalable Tree Boosting System
• SMOTE: Synthetic Minority Over-sampling Technique
• Maps of Dust IR Emission for Use in Estimation of Reddening and CMBR Foregrounds
• Astropy: A Community Python Package for Astronomy
• CANDELS: The Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey - The Hubble Space Telescope Observations, Imaging Data Products and Mosaics
• Accurate photometric redshifts for the CFHT Legacy Survey calibrated using the VIMOS VLT Deep Survey
• The COSMOS2015 Catalog: Exploring the 1<z<6 Universe with half a million galaxies
• The Hyper Suprime-Cam SSP Survey: Overview and Survey Design
• zCOSMOS: A Large VLT/VIMOS redshift survey covering 0 < z < 3 in the COSMOS field
• A new photometric technique for the joint selection of star-forming and passive galaxies at 1.4<z<2.5
• First Data Release of the Hyper Suprime-Cam Subaru Strategic Program
• The MOSFIRE Deep Evolution Field (MOSDEF) Survey: Rest-Frame Optical Spectroscopy for ~1500 H-Selected Galaxies at 1.37 < z < 3.8
• Weak Gravitational Lensing with COSMOS: Galaxy Selection and Shape Measurements
• The XMM-Newton Wide-Field Survey in the COSMOS field (XMM-COSMOS): demography and multiwavelength properties of obscured and unobscured luminous AGN
• Introduction to astroML: Machine Learning for Astrophysics
• The VIMOS Public Extragalactic Redshift Survey (VIPERS). Full spectroscopic data and auxiliary information release (PDR-2)
• Star-galaxy Classification Using Deep Convolutional Neural Networks
• The sizes, masses and specific star-formation rates of massive galaxies at 1.3<z<1.5: strong evidence in favour of evolution via minor mergers
• The FMOS-COSMOS survey of star-forming galaxies at z~1.6 III. Survey design, performance, and sample characteristics
• High velocity outflows from young star-forming galaxies in the UKIDSS Ultra-Deep Survey
• The VIMOS Ultra Deep Survey First Data Release: spectra and spectroscopic redshifts of 698 objects up to z~6 in CANDELS
• The CFHT Large Area U-band Deep Survey (CLAUDS)
• Uncovering obscured luminous AGN with WISE
• Decision Tree Classifiers for Star/Galaxy Separation
• The 0.1<z<1.65 evolution of the bright end of the [OII] luminosity function
• Obscuration in AGNs: near-infrared luminosity relations and dust colors
• A Machine Learns to Predict the Stability of Tightly Packed Planetary Systems
• A CFH12k lensing survey of X-ray luminous galaxy clusters - I: Weak lensing methodology
• Star/galaxy separation at faint magnitudes: Application to a simulated Dark Energy Survey
• Separation of pulsar signals from noise with supervised machine learning algorithms
• Star-Galaxy Classification in Multi-Band Optical Imaging
• Star-galaxy separation strategies for WISE-2MASS all-sky infrared galaxy catalogs
• UV & U-band luminosity functions from CLAUDS and HSC-SSP -- I. Using four million galaxies to simultaneously constrain the very faint and bright regimes to $z\sim 3$
• A Hybrid Ensemble Learning Approach to Star-Galaxy Classification
• Machine learning Applied to Star-Galaxy-QSO Classification and Stellar Effective Temperature Regression
• Composite Spectral Energy Distributions and Infrared-Optical Colors of Type 1 and Type 2 Quasars
• Efficient Selection of Quasar Candidates Based on Optical and Infrared Photometric Data Using Machine Learning
• The VIPERS Multi-Lambda Survey. I. UV and NIR Observations, multi-color catalogues and photometric redshifts
• Star-galaxy classification in the Dark Energy Survey Y1 dataset
• Effect of training characteristics on object classification: an application using Boosted Decision Trees
• The numbers of z~2 star-forming and passive galaxies in 2.5 square degrees of deep CFHT imaging
• The S-PLUS: a star/galaxy classification based on a Machine Learning approach

Cited by in corpus (13)

• Third Data Release of the Hyper Suprime-Cam Subaru Strategic Program
• Combining the CLAUDS & HSC-SSP surveys: U+grizy(+YJHKs) photometry and photometric redshifts for 18M galaxies in the 20 deg2 of the HSC-SSP Deep and ultraDeep fields
• Groups and protocluster candidates in the CLAUDS and HSC-SSP joint deep surveys
• Gradient boosting decision trees classification of blazars of uncertain type in the fourth Fermi-LAT catalog
• Quasar and galaxy classification using Gaia EDR3 and CatWise2020
• Estimation of Photometric Redshifts. I. Machine Learning Inference for Pan-STARRS1 Galaxies Using Neural Networks
• Photometric Selection of type 1 Quasars in the XMM-LSS Field with Machine Learning and the Disk-Corona Connection
• Exploring the Dependence of Gas Cooling and Heating Functions on the Incident Radiation Field with Machine Learning
• The viewing angle in AGN SED models, a data-driven analysis
• Improved source classification and performance analysis using Gaia DR3
• Mimicking the halo-galaxy connection using machine learning
• The miniJPAS survey quasar selection I: Mock catalogues for classification
• Identifying AGN host galaxies by Machine Learning with HSC+WISE