Auto-identification of unphysical source reconstructions in strong gravitational lens modelling

Abstract

With the advent of next-generation surveys and the expectation of discovering huge numbers of strong gravitational lens systems, much effort is being invested into developing automated procedures for handling the data. The several orders of magnitude increase in the number of strong galaxy-galaxy lens systems is an insurmountable challenge for traditional modelling techniques. Whilst machine learning techniques have dramatically improved the efficiency of lens modelling, parametric modelling of the lens mass profile remains an important tool for dealing with complex lensing systems. In particular, source reconstruction methods are necessary to cope with the irregular structure of high-redshift sources. In this paper, we consider a Convolutional Neural Network (CNN) that analyses the outputs of semi-analytic methods which parametrically model the lens mass and linearly reconstruct the source surface brightness distribution. We show the unphysical source reconstructions that arise as a result of incorrectly initialised lens models can be effectively caught by our CNN. Furthermore, the CNN predictions can be used to automatically re-initialise the parametric lens model, avoiding unphysical source reconstructions. The CNN, trained on reconstructions of lensed Sérsic sources, accurately classifies source reconstructions of the same type with a precision $P>0.99$ and recall $R>0.99$. The same CNN, without re-training, achieves $P=0.89$ and $R=0.89$ when classifying source reconstructions of more complex lensed HUDF sources. Using the CNN predictions to re-initialise the lens modelling procedure, we achieve a 69 per cent decrease in the occurrence of unphysical source reconstructions. This combined CNN and parametric modelling approach can greatly improve the automation of lens modelling.

arXiv comment13 pages, 13 figures, accepted for publication in MNRAS

References in corpus (45)

• Planck 2015 results. XIII. Cosmological parameters
• MultiNest: an efficient and robust Bayesian inference tool for cosmology and particle physics
• H0LiCOW XIII. A 2.4% measurement of $H_0$ from lensed quasars: $5.3\sigma$ tension between early and late-Universe probes
• The Hubble Ultra Deep Field
• The Sloan Lens ACS Survey. III - The Structure and Formation of Early-type Galaxies and their Evolution since z~1
• The Sloan Lens ACS Survey. I. A Large Spectroscopically Selected Sample of Massive Early-Type Lens Galaxies
• The Sloan Lens ACS Survey. V. The Full ACS Strong-Lens Sample
• TDCOSMO IV: Hierarchical time-delay cosmography -- joint inference of the Hubble constant and galaxy density profiles
• The population of galaxy-galaxy strong lenses in forthcoming optical imaging surveys
• Bayesian Strong Gravitational-Lens Modeling on Adaptive Grids: Objective Detection of Mass Substructure in Galaxies
• Fast Automated Analysis of Strong Gravitational Lenses with Convolutional Neural Networks
• A Bayesian analysis of regularised source inversions in gravitational lensing
• The Structure and Dynamics of Luminous and Dark Matter in the Early-Type Lens Galaxy of 0047-281 at z=0.485
• Semi-linear gravitational lens inversion
• The SL2S Galaxy-scale Lens Sample. IV. The dependence of the total mass density profile of early-type galaxies on redshift, stellar mass, and size
• The BOSS Emission-Line Lens Survey (BELLS). I. A large spectroscopically selected sample of Lens Galaxies at redshift ~ 0.5
• Resolved Spectroscopy of Gravitationally-Lensed Galaxies: Recovering Coherent Velocity Fields in Sub-Luminous z~2-3 Galaxies
• The CFHTLS Strong Lensing Legacy Survey: I. Survey overview and T0002 release sample
• A dynamically cold disk galaxy in the early Universe
• Gravitational lens modeling with basis sets
• Uncertainties in Parameters Estimated with Neural Networks: Application to Strong Gravitational Lensing
• AutoLens: Automated Modeling of a Strong Lens's Light, Mass and Source
• The Emission Line Properties of Gravitationally-lensed 1.5 < z < 5 Galaxies
• RingFinder: automated detection of galaxy-scale gravitational lenses in ground-based multi-filter imaging data
• Statistics of mass substructure from strong gravitational lensing: quantifying the mass fraction and mass function
• Survey of Gravitationally-lensed Objects in HSC Imaging (SuGOHI). I. Automatic search for galaxy-scale strong lenses
• Revealing the complex nature of the strong gravitationally lensed system H-ATLAS J090311.6+003906 using ALMA
• A Spatially Resolved Map of the Kinematics, Star-Formation and Stellar Mass Assembly in a Star-Forming Galaxy at z=4.9
• Constraining sterile neutrino cosmologies with strong gravitational lensing observations at redshift z~0.2
• Adaptive Semi-linear Inversion of Strong Gravitational Lens Imaging
• Automated detection of galaxy-scale gravitational lenses in high resolution imaging data
• Arcfinder: An algorithm for the automatic detection of gravitational arcs
• Identifying Strong Lenses with Unsupervised Machine Learning using Convolutional Autoencoder
• VDES J2325-5229 a z=2.7 gravitationally lensed quasar discovered using morphology independent supervised machine learning
• Data-Driven Reconstruction of Gravitationally Lensed Galaxies using Recurrent Inference Machines
• The Sloan Lens ACS Survey. XI. Beyond Hubble resolution: size, luminosity and stellar mass of compact lensed galaxies at intermediate redshift
• VIS: the visible imager for Euclid
• Cosmic Evolution of Virial and Stellar Mass in Massive Early-Type Galaxies
• Modelling high resolution ALMA observations of strongly lensed highly star forming galaxies detected by Herschel
• The use of convolutional neural networks for modelling large optically-selected strong galaxy-lens samples
• A Neural Network Gravitational Arc Finder based on the Mediatrix filamentation Method
• PICS: Simulations of Strong Gravitational Lensing in Galaxy Clusters
• A seven square degrees survey for galaxy-scale gravitational lenses with the HST imaging archive
• A search for edge-on galaxy lenses in the CFHT Legacy Survey
• On the use of shapelets in modelling resolved, gravitationally lensed images

Cited by in corpus (14)

• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• Automated galaxy-galaxy strong lens modelling: no lens left behind
• Systematic errors induced by the elliptical power-law model in galaxy-galaxy strong lens modeling
• Testing strong lensing subhalo detection with a cosmological simulation
• A machine learning based approach to gravitational lens identification with the International LOFAR Telescope
• HOLISMOKES -- IX. Neural network inference of strong-lens parameters and uncertainties from ground-based images
• The COSMOS-Web Lens Survey (COWLS) I: Discovery of >100 high redshift strong lenses in contiguous JWST imaging
• AI-driven spatio-temporal engine for finding gravitationally lensed type Ia supernovae
• Forklens: Accurate weak-lensing shear measurement with deep learning
• Exploiting the diversity of modeling methods to probe systematic biases in strong lensing analyses
• Unveiling a 36 Billion Solar Mass Black Hole at the Centre of the Cosmic Horseshoe Gravitational Lens
• Probing General Relativity in galactic scales at z $\sim 0.3$
• Systematics in ETG Mass Profile Modelling: Strong Lensing & Stellar Dynamics
• Using Strong Lensing to Detect Subhalos with Steep Inner Density Profiles