Quantifying the structure of strong gravitational lens potentials with uncertainty-aware deep neural networks

Abstract

Gravitational lensing is a powerful tool for constraining substructure in the mass distribution of galaxies, be it from the presence of dark matter sub-halos or due to physical mechanisms affecting the baryons throughout galaxy evolution. Such substructure is hard to model and is either ignored by traditional, smooth modelling, approaches, or treated as well-localized massive perturbers. In this work, we propose a deep learning approach to quantify the statistical properties of such perturbations directly from images, where only the extended lensed source features within a mask are considered, without the need of any lens modelling. Our training data consist of mock lensed images assuming perturbing Gaussian Random Fields permeating the smooth overall lens potential, and, for the first time, using images of real galaxies as the lensed source. We employ a novel deep neural network that can handle arbitrary uncertainty intervals associated with the training dataset labels as input, provides probability distributions as output, and adopts a composite loss function. The method succeeds not only in accurately estimating the actual parameter values, but also reduces the predicted confidence intervals by 10 per cent in an unsupervised manner, i.e., without having access to the actual ground truth values. Our results are invariant to the inherent degeneracy between mass perturbations in the lens and complex brightness profiles for the source. Hence, we can quantitatively and robustly quantify the smoothness of the mass density of thousands of lenses, including confidence intervals, and provide a consistent ranking for follow-up science.

arXiv comment13 pages, 10 figures, accepted for publication in MNRAS

References in corpus (40)

• Seven-Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Cosmological Interpretation
• Challenges of Big Data Analysis
• Small-Scale Challenges to the $\Lambda$CDM Paradigm
• H0LiCOW XIII. A 2.4% measurement of $H_0$ from lensed quasars: $5.3\sigma$ tension between early and late-Universe probes
• The Sloan Lens ACS Survey. X. Stellar, Dynamical, and Total Mass Correlations of Massive Early-type Galaxies
• 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. IV: the mass density profile of early-type galaxies out to 100 effective radii
• Strong Lensing by Galaxies
• The Sloan Lens ACS Survey. V. The Full ACS Strong-Lens Sample
• The Structure & Dynamics of Massive Early-type Galaxies: On Homology, Isothermality and Isotropy inside one Effective Radius
• Gravitational detection of a low-mass dark satellite at cosmological distance
• The population of galaxy-galaxy strong lenses in forthcoming optical imaging surveys
• Optimization under Uncertainty in the Era of Big Data and Deep Learning: When Machine Learning Meets Mathematical Programming
• Detection of lensing substructure using ALMA observations of the dusty galaxy SDP.81
• Detection of a Dark Substructure through Gravitational Imaging
• Fast Automated Analysis of Strong Gravitational Lenses with Convolutional Neural Networks
• Two-dimensional kinematics of SLACS lenses: III. Mass structure and dynamics of early-type lens galaxies beyond z ~ 0.1
• Cosmology from gravitational lens time delays and Planck data
• TDCOSMO. I. An exploration of systematic uncertainties in the inference of $H_0$ from time-delay cosmography
• 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
• Gravitational probes of dark matter physics
• The Strong Gravitational Lens Finding Challenge
• Detection of substructure with adaptive optics integral field spectroscopy of the gravitational lens B1422+231
• Surveying the reach and maturity of machine learning and artificial intelligence in astronomy
• Lensing substructure quantification in RXJ1131-1231: A 2 keV lower bound on dark matter thermal relic mass
• The BOSS Emission-Line Lens Survey. IV. : Smooth Lens Models for the BELLS GALLERY Sample
• Constraints on the identity of the dark matter from strong gravitational lenses
• Super-resolving distant galaxies with gravitational telescopes: Keck-LGSAO and Hubble imaging of the lens system SDSSJ0737+3216
• Mining for Dark Matter Substructure: Inferring subhalo population properties from strong lenses with machine learning
• Fermi-LAT Observations of the LIGO event GW150914
• Substructure in the lens HE 0435-1223
• SHARP - IV. An apparent flux ratio anomaly resolved by the edge-on disc in B0712+472
• Data-Driven Reconstruction of Gravitationally Lensed Galaxies using Recurrent Inference Machines
• Dark matter contraction and stellar-mass-to-light ratio gradients in massive early-type galaxies
• Detection of Substructure in the Gravitationally Lensed Quasar MG0414+0534 using Mid-Infrared and Radio VLBI Observations
• The use of convolutional neural networks for modelling large optically-selected strong galaxy-lens samples
• Deep Learning the Morphology of Dark Matter Substructure
• The inner mass power spectrum of galaxies using strong gravitational lensing: beyond linear approximation
• Differentiable Strong Lensing: Uniting Gravity and Neural Nets through Differentiable Probabilistic Programming

Cited by in corpus (12)

• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• Sensitivity of strong lensing observations to dark matter substructure: a case study with Euclid
• Using wavelets to capture deviations from smoothness in galaxy-scale strong lenses
• Strong lens modelling: comparing and combining Bayesian neural networks and parametric profile fitting
• Strong Lensing by Galaxies
• Estimating the warm dark matter mass from strong lensing images with truncated marginal neural ratio estimation
• The Very Knotty Lenser: exploring the role of regularization in source and potential reconstructions using Gaussian Process Regression
• Modeling lens potentials with continuous neural fields in galaxy-scale strong lenses
• Simulating time-varying strong lenses
• Interlopers speak out: studying the dark universe using small-scale lensing anisotropies
• Anisotropic strong lensing as a probe of dark matter self-interactions
• Measuring the Substructure Mass Power Spectrum of 23 SLACS Strong Galaxy-Galaxy Lenses with Convolutional Neural Networks