Structured Variational Inference for Simulating Populations of Radio Galaxies

Abstract

We present a model for generating postage stamp images of synthetic Fanaroff-Riley Class I and Class II radio galaxies suitable for use in simulations of future radio surveys such as those being developed for the Square Kilometre Array. This model uses a fully-connected neural network to implement structured variational inference through a variational auto-encoder and decoder architecture. In order to optimise the dimensionality of the latent space for the auto-encoder we introduce the radio morphology inception score (RAMIS), a quantitative method for assessing the quality of generated images, and discuss in detail how data pre-processing choices can affect the value of this measure. We examine the 2-dimensional latent space of the VAEs and discuss how this can be used to control the generation of synthetic populations, whilst also cautioning how it may lead to biases when used for data augmentation.

arXiv comment20 pages, 20 figures, accepted MNRAS

References in corpus (49)

• Denoising Diffusion Probabilistic Models
• Going Deeper with Convolutions
• On the fundamental dichotomy in the local radio-AGN population: accretion, evolution, and host galaxy properties
• Autoencoding beyond pixels using a learned similarity metric
• The LOFAR Two-metre Sky Survey - II. First data release
• GaLactic and Extragalactic All-sky Murchison Widefield Array (GLEAM) survey I: A low-frequency extragalactic catalogue
• EMU: Evolutionary Map of the Universe
• A 20cm VLA Survey of Abell Clusters of Galaxies VI. Radio/Optical Luminosity Functions
• Stochastic Video Generation with a Learned Prior
• Composable Effects for Flexible and Accelerated Probabilistic Programming in NumPyro
• CAN: Creative Adversarial Networks, Generating "Art" by Learning About Styles and Deviating from Style Norms
• Revisiting the Fanaroff-Riley dichotomy and radio-galaxy morphology with the LOFAR Two-Metre Sky Survey (LoTSS)
• Wasserstein Auto-Encoders
• Three-dimensional Relativistic MHD Simulations of Active Galactic Nuclei Jets: Magnetic Kink Instability and Fanaroff-Riley Dichotomy
• Classifying Radio Galaxies with Convolutional Neural Network
• Luminosity function, sizes and FR dichotomy of radio-loud AGN
• Radio Galaxy Zoo: host galaxies and radio morphologies derived from visual inspection
• Particle content, radio-galaxy morphology and jet power: all radio-loud AGN are not equal
• The relation between morphology, accretion modes and environmental factors in local radio AGN
• Radio Galaxy Zoo: ClaRAN - A Deep Learning Classifier for Radio Morphologies
• Hyperspherical Variational Auto-Encoders
• FRICAT: A FIRST catalog of FRI radio galaxies
• Radio Galaxy Zoo: Compact and extended radio source classification with deep learning
• The nuclear properties and extended morphologies of powerful radio galaxies: the roles of host galaxy and environment
• Simulation of Astronomical Images from Optical Survey Telescopes using a Comprehensive Photon Monte Carlo Approach
• The MeerKAT International GHz Tiered Extragalactic Exploration (MIGHTEE) Survey
• Morphological classification of radio galaxies: Capsule Networks versus Convolutional Neural Networks
• Variational Auto-encoded Deep Gaussian Processes
• Conditional Flow Variational Autoencoders for Structured Sequence Prediction
• CVAE-GAN: Fine-Grained Image Generation through Asymmetric Training
• Radio Galaxy Zoo: Machine learning for radio source host galaxy cross-identification
• ControlVAE: Controllable Variational Autoencoder
• Radio Galaxy Zoo: Unsupervised Clustering of Convolutionally Auto-encoded Radio-astronomical Images
• AstroVaDEr: Astronomical Variational Deep Embedder for Unsupervised Morphological Classification of Galaxies and Synthetic Image Generation
• The CoNFIG Catalogue - II. Comparison of Space Densities in the FR Dichotomy
• Transfer learning for radio galaxy classification
• Deep Learning Based Detection of Cosmological Diffuse Radio Sources
• 3D Relativistic MHD numerical simulations of X-shaped radio sources
• Semi-Amortized Variational Autoencoders
• A Conditional Variational Framework for Dialog Generation
• The Combined NVSS-FIRST Galaxies (CoNFIG) Sample - I. Sample Definition, Classification and Evolution
• Geodesics of learned representations
• Preventing Posterior Collapse with delta-VAEs
• Accelerated Discovery of Sustainable Building Materials
• VAE/WGAN-Based Image Representation Learning For Pose-Preserving Seamless Identity Replacement In Facial Images
• A Search for Extended Radio Sources in 1.3 sr of the VLA Sky Survey (VLASS)
• CDVAE: Co-embedding Deep Variational Auto Encoder for Conditional Variational Generation
• Morphological classification of radio sources for galaxy evolution and cosmology with SKA-MID
• Learning with Succinct Common Representation Based on Wyner's Common Information

Cited by in corpus (10)

• Euclid preparation: XIII. Forecasts for galaxy morphology with the Euclid Survey using Deep Generative Models
• Variational Inference as an alternative to MCMC for parameter estimation and model selection
• Quantifying Uncertainty in Deep Learning Approaches to Radio Galaxy Classification
• Morphological Classification of Radio Galaxies with wGAN-supported Augmentation
• Investigation of stellar magnetic activity using variational autoencoder based on low-resolution spectroscopic survey
• Simulating images of radio galaxies with diffusion models
• Radio Galaxy Zoo: Using semi-supervised learning to leverage large unlabelled data-sets for radio galaxy classification under data-set shift
• Radio Halo Detection in MWA Data using Deep Neural Networks and Generative Data Augmentation
• Weight Pruning and Uncertainty in Radio Galaxy Classification
• Quantifying Radio Source Morphology