Data-Driven Reconstruction of Gravitationally Lensed Galaxies using Recurrent Inference Machines

Abstract

We present a machine learning method for the reconstruction of the undistorted images of background sources in strongly lensed systems. This method treats the source as a pixelated image and utilizes the Recurrent Inference Machine (RIM) to iteratively reconstruct the background source given a lens model. Our architecture learns to minimize the likelihood of the model parameters (source pixels) given the data using the physical forward model (ray tracing simulations) while implicitly learning the prior of the source structure from the training data. This results in better performance compared to linear inversion methods, where the prior information is limited to the 2-point covariance of the source pixels approximated with a Gaussian form, and often specified in a relatively arbitrary manner. We combine our source reconstruction network with a convolutional neural network that predicts the parameters of the mass distribution in the lensing galaxies directly from telescope images, allowing a fully automated reconstruction of the background source images and the foreground mass distribution.

arXiv commentSubmitted to ApJ

References in corpus (31)

• Adam: A Method for Stochastic Optimization
• emcee: The MCMC Hammer
• Massive Dark-matter halos and Evolution of Early-type Galaxies to z=1
• Gravitational detection of a low-mass dark satellite at cosmological distance
• Detection of lensing substructure using ALMA observations of the dusty galaxy SDP.81
• Lenstronomy: multi-purpose gravitational lens modelling software package
• Fast Automated Analysis of Strong Gravitational Lenses with Convolutional Neural Networks
• A Bayesian analysis of regularised source inversions in gravitational lensing
• Galaxy growth in a massive halo in the first billion years of cosmic history
• Semi-linear gravitational lens inversion
• Cosmology from gravitational lens time delays and Planck data
• Gravitational Lens Models Based on Submillimeter Array Imaging of Herschel-selected Strongly Lensed Sub-millimeter Galaxies at z>1.5
• ALMA Imaging and Gravitational Lens Models of South Pole Telescope-Selected Dusty, Star-Forming Galaxies at High Redshifts
• CMU DeepLens: Deep Learning For Automatic Image-based Galaxy-Galaxy Strong Lens Finding
• Masses of galaxy clusters from gravitational lensing
• Gravitational lens modeling with basis sets
• Uncertainties in Parameters Estimated with Neural Networks: Application to Strong Gravitational Lensing
• ALMA Observations of SPT-Discovered, Strongly Lensed, Dusty, Star-Forming Galaxies
• Finding strong lenses in CFHTLS using convolutional neural networks
• A recurrent neural network for classification of unevenly sampled variable stars
• AutoLens: Automated Modeling of a Strong Lens's Light, Mass and Source
• Deep Recurrent Neural Networks for Supernovae Classification
• Mapping substructure in the HST Frontier Fields cluster lenses and in cosmological simulations
• Recurrent Inference Machines for Solving Inverse Problems
• Deep Convolutional Neural Networks as strong gravitational lens detectors
• Fast Molecular Outflow from a Dusty Star-Forming Galaxy in the Early Universe
• LensExtractor: A Convolutional Neural Network in Search of Strong Gravitational Lenses
• Testing Convolutional Neural Networks for finding strong gravitational lenses in KiDS
• Analyzing interferometric observations of strong gravitational lenses with recurrent and convolutional neural networks
• Sparse Lens Inversion Technique (SLIT): lens and source separability from linear inversion of the source reconstruction problem
• Modelling of the Complex CASSOWARY/SLUGS Gravitational Lenses

Cited by in corpus (46)

• Machine learning and the physical sciences
• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• Morpheus: A Deep Learning Framework For Pixel-Level Analysis of Astronomical Image Data
• Mining for Dark Matter Substructure: Inferring subhalo population properties from strong lenses with machine learning
• Astronomia ex machina: a history, primer, and outlook on neural networks in astronomy
• Automated galaxy-galaxy strong lens modelling: no lens left behind
• The use of convolutional neural networks for modelling large optically-selected strong galaxy-lens samples
• Line-of-sight effects in strong gravitational lensing
• Deep Learning the Morphology of Dark Matter Substructure
• HOLISMOKES -- IV. Efficient mass modeling of strong lenses through deep learning
• Direct Detection of Dark Matter Substructure in Strong Lens Images with Convolutional Neural Networks
• Baryon acoustic oscillations reconstruction using convolutional neural networks
• Strong Lensing by Galaxies
• Estimating the warm dark matter mass from strong lensing images with truncated marginal neural ratio estimation
• Strong lens modelling: comparing and combining Bayesian neural networks and parametric profile fitting
• Extracting the Subhalo Mass Function from Strong Lens Images with Image Segmentation
• The Very Knotty Lenser: exploring the role of regularization in source and potential reconstructions using Gaussian Process Regression
• A new calibration method of sub-halo orbital evolution for semi-analytic models
• A machine learning based approach to gravitational lens identification with the International LOFAR Telescope
• SLITronomy: towards a fully wavelet-based strong lensing inversion technique
• Phase Retrieval and Design with Automatic Differentiation
• Connecting optical morphology, environment, and HI mass fraction for low-redshift galaxies using deep learning
• Auto-identification of unphysical source reconstructions in strong gravitational lens modelling
• 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
• Deep learning for photoacoustic imaging: a survey
• Decoding Dark Matter Substructure without Supervision
• Kernel Phase and Coronagraphy with Automatic Differentiation
• Machine Learning and Cosmology
• Quantifying the structure of strong gravitational lens potentials with uncertainty-aware deep neural networks
• Strong-lensing source reconstruction with variationally optimised Gaussian processes
• A Framework for Obtaining Accurate Posteriors of Strong Gravitational Lensing Parameters with Flexible Priors and Implicit Likelihoods using Density Estimation
• Pixelated Reconstruction of Foreground Density and Background Surface Brightness in Gravitational Lensing Systems using Recurrent Inference Machines
• Deep learning the astrometric signature of dark matter substructure
• Targeted Likelihood-Free Inference of Dark Matter Substructure in Strongly-Lensed Galaxies
• 3D Detection and Characterisation of ALMA Sources through Deep Learning
• Deep learning inference with the Event Horizon Telescope II. The Zingularity framework for Bayesian artificial neural networks
• CosmicRIM : Reconstructing Early Universe by Combining Differentiable Simulations with Recurrent Inference Machines
• A proof-of-concept neural network for inferring parameters of a black hole from partial interferometric images of its shadow
• Joint Modeling of Quasar Variability and Accretion Disk Reprocessing using Latent Stochastic Differential Equations
• A Modular Deep Learning Pipeline for Galaxy-Scale Strong Gravitational Lens Detection and Modeling
• A compact group lens modeled with GIGA-Lens: Enhanced inference for complex systems
• Source REconstruction of Arcs behind Cluster Halos (SourceREACH): A New Source Reconstruction Algorithm Optimized for Giant Arcs and Galaxy Cluster Lenses
• Enhancing Gravitational Lens Study with Deep Learning: A Study on Effects of Dropout Regularization
• A New Method for the Reconstruction of Strongly Lensed Galaxies with Resolved Kinematics
• Differentiable probabilistic programming for strong gravitational lensing