Uncertainties in Parameters Estimated with Neural Networks: Application to Strong Gravitational Lensing

Abstract

In Hezaveh et al. 2017 we showed that deep learning can be used for model parameter estimation and trained convolutional neural networks to determine the parameters of strong gravitational lensing systems. Here we demonstrate a method for obtaining the uncertainties of these parameters. We review the framework of variational inference to obtain approximate posteriors of Bayesian neural networks and apply it to a network trained to estimate the parameters of the Singular Isothermal Ellipsoid plus external shear and total flux magnification. We show that the method can capture the uncertainties due to different levels of noise in the input data, as well as training and architecture-related errors made by the network. To evaluate the accuracy of the resulting uncertainties, we calculate the coverage probabilities of marginalized distributions for each lensing parameter. By tuning a single hyperparameter, the dropout rate, we obtain coverage probabilities approximately equal to the confidence levels for which they were calculated, resulting in accurate and precise uncertainty estimates. Our results suggest that neural networks can be a fast alternative to Monte Carlo Markov Chains for parameter uncertainty estimation in many practical applications, allowing more than seven orders of magnitude improvement in speed.

submitted to ApJL

References in corpus (2)

• Fast Automated Analysis of Strong Gravitational Lenses with Convolutional Neural Networks
• The SL2S Galaxy-scale Lens Sample. V. Dark Matter Halos and Stellar IMF of Massive Early-type Galaxies out to Redshift 0.8

Cited by in corpus (28)

• Non-parametric Star Formation History Reconstruction with Gaussian Processes I: Counting Major Episodes of Star Formation
• Surveying the reach and maturity of machine learning and artificial intelligence in astronomy
• Strong lensing time-delay cosmography in the 2020s
• Mining for Dark Matter Substructure: Inferring subhalo population properties from strong lenses with machine learning
• A Systematic Comparison of Bayesian Deep Learning Robustness in Diabetic Retinopathy Tasks
• An Ensemble of Bayesian Neural Networks for Exoplanetary Atmospheric Retrieval
• The use of convolutional neural networks for modelling large optically-selected strong galaxy-lens samples
• Direct Detection of Dark Matter Substructure in Strong Lens Images with Convolutional Neural Networks
• Translating neutron star observations to nuclear symmetry energy via artificial neural networks
• Breaking baryon-cosmology degeneracy with the electron density power spectrum
• Large-Scale Gravitational Lens Modeling with Bayesian Neural Networks for Accurate and Precise Inference of the Hubble Constant
• A machine learning based approach to gravitational lens identification with the International LOFAR Telescope
• SLITronomy: towards a fully wavelet-based strong lensing inversion technique
• Auto-identification of unphysical source reconstructions in strong gravitational lens modelling
• Constraining the Reionization History using Bayesian Normalizing Flows
• Benchmarking Bayesian Deep Learning on Diabetic Retinopathy Detection Tasks
• GaMPEN: A Machine Learning Framework for Estimating Bayesian Posteriors of Galaxy Morphological Parameters
• Deep Learning in Wide-field Surveys: Fast Analysis of Strong Lenses in Ground-based Cosmic Experiments
• Inferring subhalo effective density slopes from strong lensing observations with neural likelihood-ratio estimation
• Estimating Photometric Redshifts for Galaxies from the DESI Legacy Imaging Surveys with Bayesian Neural Networks Trained by DESI EDR
• Heteroscedastic Calibration of Uncertainty Estimators in Deep Learning
• Dark Matter Subhalos, Strong Lensing and Machine Learning
• A proof-of-concept neural network for inferring parameters of a black hole from partial interferometric images of its shadow
• Accelerating lensed quasar discovery and modeling with physics-informed variational autoencoders
• Reliable Uncertainties for Bayesian Neural Networks using Alpha-divergences
• AGNet: Weighing Black Holes with Machine Learning
• Parameters Estimation from the 21 cm signal using Variational Inference
• A New Method for the Reconstruction of Strongly Lensed Galaxies with Resolved Kinematics