Deep Convolutional Neural Networks as strong gravitational lens detectors

Abstract

Future large-scale surveys with high resolution imaging will provide us with a few $10^5$ new strong galaxy-scale lenses. These strong lensing systems however will be contained in large data amounts which are beyond the capacity of human experts to visually classify in a unbiased way. We present a new strong gravitational lens finder based on convolutional neural networks (CNNs). The method was applied to the Strong Lensing challenge organised by the Bologna Lens Factory. It achieved first and third place respectively on the space-based data-set and the ground-based data-set. The goal was to find a fully automated lens finder for ground-based and space-based surveys which minimizes human inspect. We compare the results of our CNN architecture and three new variations ("invariant" "views" and "residual") on the simulated data of the challenge. Each method has been trained separately 5 times on 17 000 simulated images, cross-validated using 3 000 images and then applied to a 100 000 image test set. We used two different metrics for evaluation, the area under the receiver operating characteristic curve (AUC) score and the recall with no false positive (${\mathrm{Recall}}_{0FP}$). For ground based data our best method achieved an AUC score of $0.977$ and a ${\mathrm{Recall}}_{0FP}$ of $0.50$. For space-based data our best method achieved an AUC score of $0.940$ and a ${\mathrm{Recall}}_{0FP}$ of $0.32$. On space-based data adding dihedral invariance to the CNN architecture diminished the overall score but achieved a higher no contamination recall. We found that using committees of 5 CNNs produce the best recall at zero contamination and consistenly score better AUC than a single CNN. We found that for every variation of our CNN lensfinder, we achieve AUC scores close to $1$ within $6\%$.

arXiv comment9 pages, accepted to A&A

References in corpus (34)

• Rotation-invariant convolutional neural networks for galaxy morphology prediction
• Understanding Deep Convolutional Networks
• The Sloan Lens ACS Survey. IV: the mass density profile of early-type galaxies out to 100 effective radii
• The Sloan Lens ACS Survey. V. The Full ACS Strong-Lens Sample
• Gravitationally lensed quasars and supernovae in future wide-field optical imaging surveys
• The initial mass function of early-type galaxies
• The Structure & Dynamics of Massive Early-type Galaxies: On Homology, Isothermality and Isotropy inside one Effective Radius
• H0LiCOW I. $H_0$ Lenses in COSMOGRAIL's Wellspring: Program Overview
• The population of galaxy-galaxy strong lenses in forthcoming optical imaging surveys
• Satellite Kinematics III: Halo Masses of Central Galaxies in SDSS
• The internal structure and formation of early-type galaxies: the gravitational--lens system MG2016+112 at z=1.004
• A Probable z~7 Galaxy Strongly Lensed by the Rich Cluster Abell 2218: Exploring the Dark Ages
• CMU DeepLens: Deep Learning For Automatic Image-based Galaxy-Galaxy Strong Lens Finding
• 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
• The internal structure of the lens PG1115+080: breaking degeneracies in the value of the Hubble Constant
• New Constraints on the Faint-end of the UV Luminosity Function at z~7-8 using the Gravitational Lensing of the Hubble Frontier Fields Cluster A2744
• The SL2S Galaxy-scale Lens Sample. V. Dark Matter Halos and Stellar IMF of Massive Early-type Galaxies out to Redshift 0.8
• Finding strong lenses in CFHTLS using convolutional neural networks
• The Emission Line Properties of Gravitationally-lensed 1.5 < z < 5 Galaxies
• Space Warps II. New Gravitational Lens Candidates from the CFHTLS Discovered through Citizen Science
• The Baryon Fractions and Mass-to-Light Ratios of Early-Type Galaxies
• COSMOGRAIL: the COSmological MOnitoring of GRAvItational Lenses XV. Assessing the achievability and precision of time-delay measurements
• Ideas for Citizen Science in Astronomy
• GLAMER Part I: A Code for Gravitational Lensing Simulations with Adaptive Mesh Refinement
• MOKA: a new tool for Strong Lensing Studies
• Limits on the power-law mass and luminosity density profiles of elliptical galaxies from gravitational lensing systems
• Automatic detection of arcs and arclets formed by gravitational lensing
• Strong Gravitational Lensing and the Stellar IMF of Early-type Galaxies
• A PCA-based automated finder for galaxy-scale strong lenses
• Constraining the low-mass end of the Initial Mass Function with Gravitational Lensing
• A seven square degrees survey for galaxy-scale gravitational lenses with the HST imaging archive
• The PCA Lens-Finder: application to CFHTLS
• Probing a massive radio galaxy with gravitational lensing

Cited by in corpus (59)

• The Strong Gravitational Lens Finding Challenge
• Radio Galaxy Zoo: ClaRAN - A Deep Learning Classifier for Radio Morphologies
• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• Application of Deep Learning methods to analysis of Imaging Atmospheric Cherenkov Telescopes data
• Reconstructing Functions and Estimating Parameters with Artificial Neural Networks: A Test with the Hubble Parameter and SNe Ia
• HOLISMOKES -- II. Identifying galaxy-scale strong gravitational lenses in Pan-STARRS using convolutional neural networks
• Data-Driven Reconstruction of Gravitationally Lensed Galaxies using Recurrent Inference Machines
• Morphological classification of galaxies with deep learning: comparing 3-way and 4-way CNNs
• Testing Convolutional Neural Networks for finding strong gravitational lenses in KiDS
• The use of convolutional neural networks for modelling large optically-selected strong galaxy-lens samples
• Detecting solar system objects with convolutional neural networks
• Deep Learning for Strong Lensing Search: Tests of the Convolutional Neural Networks and New Candidates from KiDS DR3
• 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
• A deep learning model to emulate simulations of cosmic reionization
• ECoPANN: A Framework for Estimating Cosmological Parameters using Artificial Neural Networks
• Hubble Asteroid Hunter: II. Identifying strong gravitational lenses in HST images with crowdsourcing
• Strong lens modelling: comparing and combining Bayesian neural networks and parametric profile fitting
• The evolution of barred galaxies in the EAGLE simulations
• Recovering the CMB Signal with Machine Learning
• SILVERRUSH X: Machine Learning-Aided Selection of $9318$ LAEs at $z=2.2$, $3.3$, $4.9$, $5.7$, $6.6$, and $7.0$ from the HSC SSP and CHORUS Survey Data
• On the detection and precise localisation of merging black holes events through strong gravitational lensing
• A machine learning based approach to gravitational lens identification with the International LOFAR Telescope
• A thorough investigation of the prospects of eLISA in addressing the Hubble tension: Fisher Forecast, MCMC and Machine Learning
• Inferring galaxy dark halo properties from visible matter with Machine Learning
• The galaxy morphology-density relation in the EAGLE simulation
• Likelihood-free Inference with Mixture Density Network
• Neural network-based anomaly detection for high-resolution X-ray spectroscopy
• Deep-CEE I: Fishing for Galaxy Clusters with Deep Neural Nets
• HOLISMOKES -- IX. Neural network inference of strong-lens parameters and uncertainties from ground-based images
• Finding Strong Gravitational Lenses Through Self-Attention
• A neural network classifier for electron identification on the DAMPE experiment
• Advances in Deep Space Exploration via Simulators & Deep Learning
• HOLISMOKES -- X. Comparison between neural network and semi-automated traditional modeling of strong lenses
• Deep-Learning Study of the 21cm Differential Brightness Temperature During the Epoch of Reionization
• 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
• Finding quadruply imaged quasars with machine learning. I. Methods
• Identification of Grand-design and Flocculent Spirals from SDSS using Convolutional Neural network
• A Bayesian Approach to Strong Lens Finding in the Era of Wide-area Surveys
• Shedding Light on Low Surface Brightness Galaxies in Dark Energy Survey with Transformers
• TEGLIE: Transformer encoders as strong gravitational lens finders in KiDS
• ULISSE: A Tool for One-shot Sky Exploration and its Application to Active Galactic Nuclei Detection
• Weak-lensing Mass Reconstruction of Galaxy Clusters with Convolutional Neural Network
• A Comparative Study of Convolutional Neural Networks for the Detection of Strong Gravitational Lensing
• DeepGraviLens: a Multi-Modal Architecture for Classifying Gravitational Lensing Data
• Anisotropic strong lensing as a probe of dark matter self-interactions
• Accurately Estimating Redshifts from CSST Slitless Spectroscopic Survey using Deep Learning
• Probing Neural Networks for the Gamma/Hadron Separation of the Cherenkov Telescope Array
• Dark Matter Subhalos, Strong Lensing and Machine Learning
• HOLISMOKES XV. Search for strong gravitational lenses combining ground-based and space-based imaging
• A proof-of-concept neural network for inferring parameters of a black hole from partial interferometric images of its shadow
• Measuring the Substructure Mass Power Spectrum of 23 SLACS Strong Galaxy-Galaxy Lenses with Convolutional Neural Networks
• Transformers as Strong Lens Detectors- From Simulation to Surveys
• Classifying merger stages with adaptive deep learning and cosmological hydrodynamical simulations
• Gaia GraL: Gaia gravitational lens systems IX. Using XGBoost to explore the Gaia Focused Product Release GravLens catalogue
• ULISSE: Determination of star-formation rate and stellar mass based on the one-shot galaxy imaging technique
• Construction of a far ultraviolet all sky map from an incomplete survey: Application of a deep learning algorithm
• Breaking the degeneracy in stellar spectral classification from single wide-band images