Holismokes

Abstract

Modeling of strong gravitational lenses is a necessity for further applications in astrophysics and cosmology. With the large number of detections in current and upcoming surveys, such as the Rubin Legacy Survey of Space and Time (LSST), it is pertinent to investigate automated and fast analysis techniques beyond the traditional and time-consuming Markov chain Monte Carlo sampling methods. Building upon our (simple) convolutional neural network (CNN), we present here another CNN, specifically a residual neural network (ResNet), that predicts the five mass parameters of a singular isothermal ellipsoid (SIE) profile (lens center x and y, ellipticity ex and ey, Einstein radius θE) and the external shear (γext, 1, γext, 2) from ground-based imaging data. In contrast to our previous CNN, this ResNet further predicts the 1σ uncertainty for each parameter. To train our network, we use our improved pipeline to simulate lens images using real images of galaxies from the Hyper Suprime-Cam Survey (HSC) and from the Hubble Ultra Deep Field as lens galaxies and background sources, respectively. We find very good recoveries overall for the SIE parameters, especially for the lens center in comparison to our previous CNN, while significant differences remain in predicting the external shear. From our multiple tests, it appears that most likely the low ground-based image resolution is the limiting factor in predicting the external shear. Given the run time of milli-seconds per system, our network is perfectly suited to quickly predict the next appearing image and time delays of lensed transients. Therefore, we use the network-predicted mass model to estimate these quantities and compare to those values obtained from our simulations. Unfortunately, the achieved precision allows only a first-order estimate of time delays on real lens systems and requires further refinement through follow-up modeling. Nonetheless, our ResNet is able to predict the SIE and shear parameter values in fractions of a second on a single CPU, meaning that we are able to efficiently process the huge amount of galaxy-scale lenses expected in the near future.

lemon: LEns MOdelling with Neural networks – I. Automated modelling of strong gravitational lenses with Bayesian Neural Networks

“…(ii) For the algorithms based on a BNN (Perreault Le v asseur et al 2017 ; Bom et al 2019 ;Pearson et al 2021 ;Schuldt et al 2023 ), one of the main differences is the calibration performed on the uncertainties predicted by the network. While the other codes changed the keeprate of the dropout layers to achieve a Gaussian co v erage of the testset, the LEMON algorithm is the only one to employ the a posteriori procedure discussed in Section 4.…”

“…While the other codes changed the keeprate of the dropout layers to achieve a Gaussian co v erage of the testset, the LEMON algorithm is the only one to employ the a posteriori procedure discussed in Section 4. 2023) Bom et al 2019 ;Schuldt et al 2023 ). These differences strongly impact the dimension of the data sets required to successfully train the algorithms and -therefore -the possibility of training them only by employing the first data provided by Euclid or LSST.…”

“…Hezaveh, Perreault Le v asseur & Marshall 2017 ;Pearson, Li & Dye 2019 ;Schuldt et al 2021 ) or Bayesian Neural Networks (BNNs; e.g. Perreault Le v asseur, Hezaveh & Wechsler 2017 ;Bom et al 2019 ;Park et al 2021 ;Pearson et al 2021 ;Schuldt et al 2023 ): supervised-learning algorithms able to extract meaningful features from images and to convert them into useful parameters (LeCun, Bengio & Hinton 2015 ;Charnock, Perreault-Le v asseur & Lanusse 2020 ). In this work, we focus on BNNs.…”