Strong lens modelling: comparing and combining Bayesian neural networks and parametric profile fitting

Abstract

The vast quantity of strong galaxy-galaxy gravitational lenses expected by future large-scale surveys necessitates the development of automated methods to efficiently model their mass profiles. For this purpose, we train an approximate Bayesian convolutional neural network (CNN) to predict mass profile parameters and associated uncertainties, and compare its accuracy to that of conventional parametric modelling for a range of increasingly complex lensing systems. These include standard smooth parametric density profiles, hydrodynamical EAGLE galaxies and the inclusion of foreground mass structures, combined with parametric sources and sources extracted from the Hubble Ultra Deep Field. In addition, we also present a method for combining the CNN with traditional parametric density profile fitting in an automated fashion, where the CNN provides initial priors on the latter's parameters. On average, the CNN achieved errors 19 $\pm$ 22 per cent lower than the traditional method's blind modelling. The combination method instead achieved 27 $\pm$ 11 per cent lower errors over the blind modelling, reduced further to 37 $\pm$ 11 per cent when the priors also incorporated the CNN-predicted uncertainties, with errors also 17 $\pm$ 21 per cent lower than the CNN by itself. While the CNN is undoubtedly the fastest modelling method, the combination of the two increases the speed of conventional fitting alone by factors of 1.73 and 1.19 with and without CNN-predicted uncertainties, respectively. This, combined with greatly improved accuracy, highlights the benefits one can obtain through combining neural networks with conventional techniques in order to achieve an efficient automated modelling approach.

arXiv comment21 pages, 19 figures, 4 tables, accepted for publication in MNRAS

References in corpus (103)

• The EAGLE project: Simulating the evolution and assembly of galaxies and their environments
• MultiNest: an efficient and robust Bayesian inference tool for cosmology and particle physics
• The EAGLE simulations of galaxy formation: calibration of subgrid physics and model variations
• 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. I. A Large Spectroscopically Selected Sample of Massive Early-Type Lens Galaxies
• New physics in light of the $H_0$ tension: an alternative view
• H0LiCOW V. New COSMOGRAIL time delays of HE0435-1223: $H_0$ to 3.8% precision from strong lensing in a flat $\Lambda$CDM model
• H0LiCOW - IX. Cosmographic analysis of the doubly imaged quasar SDSS 1206+4332 and a new measurement of the Hubble constant
• TDCOSMO IV: Hierarchical time-delay cosmography -- joint inference of the Hubble constant and galaxy density profiles
• H0LiCOW I. $H_0$ Lenses in COSMOGRAIL's Wellspring: Program Overview
• The population of galaxy-galaxy strong lenses in forthcoming optical imaging surveys
• Detection of lensing substructure using ALMA observations of the dusty galaxy SDP.81
• Lenstronomy: multi-purpose gravitational lens modelling software package
• Bayesian Strong Gravitational-Lens Modeling on Adaptive Grids: Objective Detection of Mass Substructure in Galaxies
• Does the Hubble constant tension call for new physics?
• Fast Automated Analysis of Strong Gravitational Lenses with Convolutional Neural Networks
• A Bayesian analysis of regularised source inversions in gravitational lensing
• STRIDES: a 3.9 per cent measurement of the Hubble constant from the strong lens system DES J0408-5354
• Semi-linear gravitational lens inversion
• The Superluminous Transient ASASSN-15lh as a Tidal Disruption Event from a Kerr Black Hole
• Inference of the Cold Dark Matter substructure mass function at z=0.2 using strong gravitational lenses
• CMU DeepLens: Deep Learning For Automatic Image-based Galaxy-Galaxy Strong Lens Finding
• The BOSS Emission-Line Lens Survey (BELLS). I. A large spectroscopically selected sample of Lens Galaxies at redshift ~ 0.5
• Finding Strong Gravitational Lenses in the Kilo Degree Survey with Convolutional Neural Networks
• The CFHTLS Strong Lensing Legacy Survey: I. Survey overview and T0002 release sample
• Strong gravitational lensing of explosive transients
• Gravitational lens modeling with basis sets
• A precise extragalactic test of General Relativity
• The SL2S Galaxy-scale Lens Sample. V. Dark Matter Halos and Stellar IMF of Massive Early-type Galaxies out to Redshift 0.8
• The Strong Gravitational Lens Finding Challenge
• The BOSS Emission-Line Lens Survey. II. Investigating Mass-Density Profile Evolution in the SLACS+BELLS Strong Gravitational Lens Sample
• Uncertainties in Parameters Estimated with Neural Networks: Application to Strong Gravitational Lensing
• Finding strong lenses in CFHTLS using convolutional neural networks
• AutoLens: Automated Modeling of a Strong Lens's Light, Mass and Source
• A model-independent determination of the Hubble constant from lensed quasars and supernovae using Gaussian process regression
• CosmoDC2: A Synthetic Sky Catalog for Dark Energy Science with LSST
• Low-mass halo perturbations in strong gravitational lenses at redshift z$\sim$0.5 are consistent with CDM
• Space Warps II. New Gravitational Lens Candidates from the CFHTLS Discovered through Citizen Science
• Gravitationally Lensed Quasars in Gaia: II. Discovery of 24 Lensed Quasars
• The Sloan Lens ACS Survey. XIII. Discovery of 40 New Galaxy-Scale Strong Lenses
• Revealing the complex nature of the strong gravitationally lensed system H-ATLAS J090311.6+003906 using ALMA
• The Outer Rim Simulation: A Path to Many-Core Supercomputers
• Discovering New Strong Gravitational Lenses in the DESI Legacy Imaging Surveys
• Deep Convolutional Neural Networks as strong gravitational lens detectors
• The elliptical power law profile lens
• Identifying Strong Lenses with Unsupervised Machine Learning using Convolutional Autoencoder
• The SWELLS Survey. I. A large spectroscopically selected sample of edge-on late-type lens galaxies
• Mining for Dark Matter Substructure: Inferring subhalo population properties from strong lenses with machine learning
• HOLISMOKES -- II. Identifying galaxy-scale strong gravitational lenses in Pan-STARRS using convolutional neural networks
• New high-quality strong lens candidates with deep learning in the Kilo Degree Survey
• Model-independent determination of $H_0$ and ${\Omega }_{K0}$ from strong lensing and type Ia supernovae
• A Candidate $z\sim 10$ Galaxy Strongly Lensed into a Spatially Resolved Arc
• Survey of Gravitationally-lensed Objects in HSC Imaging (SuGOHI). VI. Crowdsourced lens finding with Space Warps
• Data-Driven Reconstruction of Gravitationally Lensed Galaxies using Recurrent Inference Machines
• The Hubble constant from eight time-delay galaxy lenses
• Using Convolutional Neural Networks to identify Gravitational Lenses in Astronomical images
• The Hubble Constant determined through an inverse distance ladder including quasar time delays and Type Ia supernovae
• Cosmological model-independent constraints on spatial curvature from strong gravitational lensing and type Ia supernova observations
• Testing Star Formation Laws in a Starburst Galaxy At Redshift 3 Resolved with ALMA
• Galaxy And Mass Assembly (GAMA): The sSFR-M* relation part I - ${\sigma }_{\mathrm{sSFR}}$-M* as a function of sample, SFR indicator and environment
• Modelling high resolution ALMA observations of strongly lensed highly star forming galaxies detected by Herschel
• Hierarchical Inference With Bayesian Neural Networks: An Application to Strong Gravitational Lensing
• The use of convolutional neural networks for modelling large optically-selected strong galaxy-lens samples
• A novel approach to visibility-space modelling of interferometric gravitational lens observations at high angular resolution
• Probing the cosmic distance duality relation using time delay lenses
• The VIMOS Public Extragalactic Redshift Survey (VIPERS). The coevolution of galaxy morphology and colour to z~1
• Deep Learning for Strong Lensing Search: Tests of the Convolutional Neural Networks and New Candidates from KiDS DR3
• Deep Learning the Morphology of Dark Matter Substructure
• Direct Detection of Dark Matter Substructure in Strong Lens Images with Convolutional Neural Networks
• The first simultaneous measurement of Hubble constant and post-Newtonian parameter from Time-Delay Strong Lensing
• The Discovery of a Five-Image Lensed Quasar at z = 3.34 using PanSTARRS1 and Gaia
• A Neural Network Gravitational Arc Finder based on the Mediatrix filamentation Method
• Constraining the microlensing effect on time delays with new time-delay prediction model in $H_0$ measurements
• Differentiable Strong Lensing: Uniting Gravity and Neural Nets through Differentiable Probabilistic Programming
• MaxiMask and MaxiTrack: two new tools for identifying contaminants in astronomical images using convolutional neural networks
• The Completed SDSS-IV extended Baryon Oscillation Spectroscopic Survey: A Catalogue of Strong Galaxy-Galaxy Lens Candidates
• PICS: Simulations of Strong Gravitational Lensing in Galaxy Clusters
• Quantifying Non-parametric Structure of High-redshift Galaxies with Deep Learning
• A Triple Rollover: A third multiply-imaged source at z~6 behind the Jackpot gravitational lens
• SEAGLE - I: A pipeline for simulating and modeling strong lenses from cosmological hydrodynamic simulations
• Large-Scale Gravitational Lens Modeling with Bayesian Neural Networks for Accurate and Precise Inference of the Hubble Constant
• SDSS J0909$+$4449: A Large-separation Strongly Lensed Quasar at $z\sim 2.8$ with Three Images
• The Discovery of a Gravitationally Lensed Supernova Ia at Redshift 2.22
• Resolved Molecular Gas and Star Formation Properties of the Strongly Lensed z=2.26 Galaxy SDSS J0901+1814
• A magnified view of circumnuclear star formation and feedback around an AGN at z=2.6
• Analyzing interferometric observations of strong gravitational lenses with recurrent and convolutional neural networks
• The Hubble constant tension with next-generation galaxy surveys
• A model independent comparison of supernova and strong lensing cosmography: implications for the Hubble constant tension
• Automated Lensing Learner: Automated Strong Lensing Identification with a Computer Vision Technique
• COOL-LAMPS I. An Extraordinarily Bright Lensed Galaxy at Redshift 5.04
• H-ATLAS/GAMA: Quantifying the Morphological Evolution of the Galaxy Population Using Cosmic Calorimetry
• Be It Unresolved: Measuring Time Delays from Lensed Supernovae
• An extended catalog of galaxy-galaxy strong gravitational lenses discovered in DES using convolutional neural networks
• Models of gravitational lens candidates from Space Warps CFHTLS
• Quantifying the structure of strong gravitational lens potentials with uncertainty-aware deep neural networks
• A bright lensed galaxy at $z=5.4$ with strong Ly$\alpha$ emission
• Deep Learning in Wide-field Surveys: Fast Analysis of Strong Lenses in Ground-based Cosmic Experiments
• Testing the Equivalence Principle with Strong Lensing Time Delay Variations
• Predicting galaxy spectra from images with hybrid convolutional neural networks
• Observing the earliest moments of supernovae using strong gravitational lenses
• ALMA 50-parsec resolution imaging of jet-ISM interaction in the lensed quasar MGJ0414+0534
• Hunting for Dark Matter Subhalos in Strong Gravitational Lensing with Neural Networks

Cited by in corpus (24)

• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• Strong lensing time-delay cosmography in the 2020s
• STRIDES: Automated uniform models for 30 quadruply imaged quasars
• Automated galaxy-galaxy strong lens modelling: no lens left behind
• From Images to Dark Matter: End-To-End Inference of Substructure From Hundreds of Strong Gravitational Lenses
• Quantifying Non-parametric Structure of High-redshift Galaxies with Deep Learning
• GAlaxy Light profile convolutional neural NETworks (GaLNets). I. fast and accurate structural parameters for billion galaxy samples
• 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
• RXJ0437+00: Constraining Dark Matter with Exotic Gravitational Lenses
• LeMoN: Lens Modelling with Neural networks -- I. Automated modelling of strong gravitational lenses with Bayesian Neural Networks
• HOLISMOKES -- XI. Evaluation of supervised neural networks for strong-lens searches in ground-based imaging surveys
• AI-driven spatio-temporal engine for finding gravitationally lensed type Ia supernovae
• Exploiting the diversity of modeling methods to probe systematic biases in strong lensing analyses
• Forklens: Accurate weak-lensing shear measurement with deep learning
• Developing a Victorious Strategy to the Second Strong Gravitational Lensing Data Challenge
• HOLISMOKES XIII: Strong-lens candidates at all mass scales and their environments from the Hyper-Suprime Cam and deep learning
• Lens Modeling of STRIDES Strongly Lensed Quasars using Neural Posterior Estimation
• Measuring the Substructure Mass Power Spectrum of 23 SLACS Strong Galaxy-Galaxy Lenses with Convolutional Neural Networks
• Harvesting the Lyα forest with convolutional neural networks
• Accelerating lensed quasar discovery and modeling with physics-informed variational autoencoders
• A compact group lens modeled with GIGA-Lens: Enhanced inference for complex systems
• Deep inference of simulated strong lenses in ground-based surveys
• Enhancing Gravitational Lens Study with Deep Learning: A Study on Effects of Dropout Regularization