Quantifying Non-parametric Structure of High-redshift Galaxies with Deep Learning

Abstract

At high redshift, due to both observational limitations and the variety of galaxy morphologies in the early universe, measuring galaxy structure can be challenging. Non-parametric measurements such as the CAS system have thus become an important tool due to both their model-independent nature and their utility as a straightforward computational process. Recently, convolutional neural networks (CNNs) have been shown to be adept at image analysis, and are beginning to supersede traditional measurements of visual morphology and model-based structural parameters. In this work, we take a further step by extending CNNs to measure well known non-parametric structural quantities: concentration ($C$) and asymmetry ($A$). We train CNNs to predict $C$ and $A$ from individual images of $\sim 150,000$ galaxies at $0<z<7$ in the CANDELS fields, using Bayesian hyperparameter optimisation to select suitable network architectures. Our resulting networks accurately reproduce measurements compared with standard algorithms. Furthermore, using simulated images, we show that our networks are more stable than the standard algorithms at low signal-to-noise. While both approaches suffer from similar systematic biases with redshift, these remain small out to $z\sim 7$. Once trained, measurements with our networks are $>10^3$ times faster than previous methods. Our approach is thus able to reproduce standard measures of non-parametric morphologies and shows the potential of employing neural networks to provide superior results in substantially less time. This will be vital for making best use of the large and complex datasets provided by upcoming galaxy surveys, such as Euclid and Rubin-LSST.

arXiv comment18 pages, 13 figures, accepted for publication in ApJ

References in corpus (55)

• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Binary companions of evolved stars in APOGEE DR14: Search method and catalog of ~5,000 companions
• Detailed Structural Decomposition of Galaxy Images
• CANDELS: The Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey
• CANDELS: The Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey - The Hubble Space Telescope Observations, Imaging Data Products and Mosaics
• Euclid Definition Study Report
• The Relationship Between Stellar Light Distributions of Galaxies and their Formation Histories
• A New Non-Parametric Approach to Galaxy Morphological Classification
• Rotation-invariant convolutional neural networks for galaxy morphology prediction
• Galaxy Zoo 2: detailed morphological classifications for 304,122 galaxies from the Sloan Digital Sky Survey
• A Catalog of Bulge+Disk Decompositions and Updated Photometry for 1.12 Million Galaxies in the Sloan Digital Sky Survey
• Galaxy morphology to I=25 mag in the Hubble Deep Field
• Strong size evolution of the most massive galaxies since z~2
• Scalable Bayesian Optimization Using Deep Neural Networks
• Size evolution of the most massive galaxies at 1.7<z<3 from GOODS NICMOS survey imaging
• Galaxy Merger Morphologies and Time-Scales from Simulations of Equal-Mass Gas-Rich Disc Mergers
• Improving galaxy morphologies for SDSS with Deep Learning
• Structural and Photometric Classification of Galaxies - I. Calibration Based on a Nearby Galaxy Sample
• A catalog of visual-like morphologies in the 5 CANDELS fields using deep-learning
• The Structures of Distant Galaxies I: Galaxy Structures and the Merger Rate to z~3 in the Hubble Ultra-Deep Field
• THE CANADA-FRANCE REDSHIFT SURVEY IX: HST Imaging of High-Redshift Field Galaxies
• MegaMorph - multi-wavelength measurement of galaxy structure: complete Sérsic profile information from modern surveys
• Galaxy Zoo: Reproducing Galaxy Morphologies Via Machine Learning
• Total Galaxy Magnitudes and Effective Radii from Petrosian Magnitudes and Radii
• A Catalog of Digital Images of 113 Nearby Galaxies
• Photometric redshifts from SDSS images using a Convolutional Neural Network
• Observational constraints on the merger history of galaxies since $z\approx 6$: Probabilistic galaxy pair counts in the CANDELS fields
• Galaxy Morphologies in the Hubble Ultra Deep Field: Dominance of Linear Structures at the Detection Limit
• The structures and total (minor + major) merger histories of massive galaxies up to z = 3 in the HST GOODS NICMOS Survey: A possible solution to the size evolution problem
• The Redshift and Mass Dependence on the Formation of The Hubble Sequence at z>1 from CANDELS/UDS
• Photometric redshift estimation via deep learning
• Classifying Radio Galaxies with Convolutional Neural Network
• Flare Statistics for Young Stars from a Convolutional Neural Network Analysis of $\text{TESS}$ Data
• A robust morphological classification of high-redshift galaxies using support vector machines on seeing limited images. I Method description
• AUTOMATED MORPHOLOGICAL CLASSIFICATION OF APM GALAXIES BY SUPERVISED ARTIFICIAL NEURAL NETWORKS
• ProFit: Bayesian Profile Fitting of Galaxy Images
• Optimising Automatic Morphological Classification of Galaxies with Machine Learning and Deep Learning using Dark Energy Survey Imaging
• Machine and Deep Learning Applied to Galaxy Morphology -- A Comparative Study
• New Image Statistics for Detecting Disturbed Galaxy Morphologies at High Redshift
• Galaxy Merger Rates up to z $\sim$ 3 using a Bayesian Deep Learning Model $-$ A Major-Merger classifier using IllustrisTNG Simulation data
• Deep learning for galaxy surface brightness profile fitting
• The Symmetry, Color, and Morphology of Galaxies
• The Structures of Distant Galaxies - II: Diverse Galaxy Structures and Local Environments at z = 4-6; Implications for Early Galaxy Assembly
• The Tumultuous Formation of the Hubble Sequence at z > 1 Examined with HST/WFC3 Observations of the Hubble Ultra Deep Field
• Accurate Identification of Galaxy Mergers with Imaging
• Photometry of high-redshift blended galaxies using deep learning
• FERENGI: Redshifting galaxies from SDSS to GEMS, STAGES and COSMOS
• The use of convolutional neural networks for modelling large optically-selected strong galaxy-lens samples
• The Morphology-Density relationship in 1<z<2 clusters
• A New Automatic Method to Identify Galaxy Mergers I. Description and Application to the STAGES Survey
• Strong lens modelling: comparing and combining Bayesian neural networks and parametric profile fitting
• Unbiased Differential Size Evolution and the Inside-Out Growth of Galaxies in the Deep CANDELS GOODS Fields at $1\le z\le 7$
• Unsupervised feature-learning for galaxy SEDs with denoising autoencoders
• Surface Brightness Evolution of Galaxies in the CANDELS GOODS Fields up to $z\sim 6$: High-z Galaxies are Unique or Remain Undetected
• The Impact of Redshift on Galaxy Morphometric Classification: case studies for SDSS, DES, LSST and HST with Morfometryka

Cited by in corpus (22)

• Euclid. I. Overview of the Euclid mission
• EPOCHS VI: The Size and Shape Evolution of Galaxies since z ~ 8 with JWST Observations
• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• Characterization of Low Surface Brightness structures in annotated deep images
• Galaxy Evolution in all Five CANDELS Fields and IllustrisTNG: Morphological, Structural, and the Major Merger Evolution to $z\sim 3$
• Strong lens modelling: comparing and combining Bayesian neural networks and parametric profile fitting
• A Robust Study of High-Redshift Galaxies: Unsupervised Machine Learning for Characterising morphology with JWST up to z ~ 8
• Redshifting galaxies from DESI to JWST CEERS: Correction of biases and uncertainties in quantifying morphology
• Galaxy morphoto-Z with neural Networks (GaZNets). I. Optimized accuracy and outlier fraction from Imaging and Photometry
• JWST's PEARLS: resolved study of the stellar and dust components in starburst galaxies at cosmic noon
• HOLISMOKES -- IX. Neural network inference of strong-lens parameters and uncertainties from ground-based images
• HOLISMOKES -- X. Comparison between neural network and semi-automated traditional modeling of strong lenses
• HOLISMOKES -- XI. Evaluation of supervised neural networks for strong-lens searches in ground-based imaging surveys
• Unveiling Galaxy Morphology through an Unsupervised-Supervised Hybrid Approach
• Lessons Learned from the Two Largest Galaxy Morphological Classification Catalogues built by Convolutional Neural Networks
• HOLISMOKES XIII: Strong-lens candidates at all mass scales and their environments from the Hyper-Suprime Cam and deep learning
• The PSF Smoothing Effect on Concentration-Related Parameters of High Redshift Galaxies in HST and JWST
• Galaxy Zoo JWST: Up to 75% of discs are featureless at $3<z<7$
• On the application of dimensionality reduction and clustering algorithms for the classification of kinematic morphologies of galaxies
• The Next Generation Virgo Cluster Survey (NGVS). XL. The Morphological Classification of Virgo Cluster Galaxies
• A Simulation Driven Deep Learning Approach for Separating Mergers and Star Forming Galaxies: The Formation Histories of Clumpy Galaxies in all the CANDELS Fields
• Quantifying the Poor Purity and Completeness of Morphological Samples Selected by Galaxy Colour