Deep Generative Models for Galaxy Image Simulations

Abstract

Image simulations are essential tools for preparing and validating the analysis of current and future wide-field optical surveys. However, the galaxy models used as the basis for these simulations are typically limited to simple parametric light profiles, or use a fairly limited amount of available space-based data. In this work, we propose a methodology based on Deep Generative Models to create complex models of galaxy morphologies that may meet the image simulation needs of upcoming surveys. We address the technical challenges associated with learning this morphology model from noisy and PSF-convolved images by building a hybrid Deep Learning/physical Bayesian hierarchical model for observed images, explicitly accounting for the Point Spread Function and noise properties. The generative model is further made conditional on physical galaxy parameters, to allow for sampling new light profiles from specific galaxy populations. We demonstrate our ability to train and sample from such a model on galaxy postage stamps from the HST/ACS COSMOS survey, and validate the quality of the model using a range of second- and higher-order morphology statistics. Using this set of statistics, we demonstrate significantly more realistic morphologies using these deep generative models compared to conventional parametric models. To help make these generative models practical tools for the community, we introduce GalSim-Hub, a community-driven repository of generative models, and a framework for incorporating generative models within the GalSim image simulation software.

14 pages, submitted to MNRAS. Comments most welcome

References in corpus (11)

• The Cosmic Evolution Survey (COSMOS) -- Overview
• Weak Gravitational Lensing with COSMOS: Galaxy Selection and Shape Measurements
• A catalogue of 2D photometric decompositions in the SDSS-DR7 spectroscopic main galaxy sample: preferred models and systematics
• The LSST DESC DC2 Simulated Sky Survey
• The Advanced Camera for Surveys General Catalog: Structural Parameters for Approximately Half a Million Galaxies
• The relationship between fine galaxy stellar morphology and star formation activity in cosmological simulations: a deep learning view
• Weak gravitational shear and flexion with polar shapelets
• Deblending galaxies with Variational Autoencoders: a joint multi-band, multi-instrument approach
• The LSST DESC Data Challenge 1: Generation and Analysis of Synthetic Images for Next Generation Surveys
• The impact of cosmic variance on simulating weak lensing surveys
• The metallicity and elemental abundance maps of kinematically atypical galaxies for constraining minor merger and accretion histories

Cited by in corpus (30)

• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• Realistic galaxy image simulation via score-based generative models
• Strong lensing in UNIONS: Toward a pipeline from discovery to modeling
• Euclid preparation: XIII. Forecasts for galaxy morphology with the Euclid Survey using Deep Generative Models
• Autoencoding Galaxy Spectra I: Architecture
• Euclid preparation. XXV. The Euclid Morphology Challenge -- Towards model-fitting photometry for billions of galaxies
• HIFlow: Generating Diverse HI Maps and Inferring Cosmology while Marginalizing over Astrophysics using Normalizing Flows
• Fast and robust Bayesian Inference using Gaussian Processes with GPry
• Realistic galaxy images and improved robustness in machine learning tasks from generative modelling
• Scalable inference with Autoregressive Neural Ratio Estimation
• HyPhy: Deep Generative Conditional Posterior Mapping of Hydrodynamical Physics
• From Images to Features: Unbiased Morphology Classification via Variational Auto-Encoders and Domain Adaptation
• Pixelated Reconstruction of Foreground Density and Background Surface Brightness in Gravitational Lensing Systems using Recurrent Inference Machines
• Survey2Survey: A deep learning generative model approach for cross-survey image mapping
• Forklens: Accurate weak-lensing shear measurement with deep learning
• 3D Detection and Characterisation of ALMA Sources through Deep Learning
• Searching for galaxy-scale strong-lenses in galaxy clusters with deep networks -- I: methodology and network performance
• Variational Inference for Deblending Crowded Starfields
• Neural Network Based Point Spread Function Deconvolution For Astronomical Applications
• Galaxy Imaging with Generative Models: Insights from a Two-Models Framework
• Measuring the Substructure Mass Power Spectrum of 23 SLACS Strong Galaxy-Galaxy Lenses with Convolutional Neural Networks
• Tackling the Problem of Distributional Shifts: Correcting Misspecified, High-Dimensional Data-Driven Priors for Inverse Problems
• Weak-lensing Mass Reconstruction of Galaxy Clusters with a Convolutional Neural Network -- II: Application to Next-Generation Wide-Field Surveys
• Denoising Diffusion Probabilistic Model for realistic and fast generated \textit{Euclid}-like data for weak lensing analysis
• MEGS: Morphological Evaluation of Galactic Structure
• Euclid preparation. LXVII. Deep learning true galaxy morphologies for weak lensing shear bias calibration
• Investigation on deep learning-based galaxy image translation models
• Neural Posterior Estimation for Cataloging Astronomical Images with Spatially Varying Backgrounds and Point Spread Functions
• Can AI Dream of Unseen Galaxies? Conditional Diffusion Model for Galaxy Morphology Augmentation
• Comparison of Graphcore IPUs and Nvidia GPUsfor cosmology applications