Holismokes

Schuldt, S.; Suyu, S. H.; Cañameras, R.; Shu, Y.; Taubenberger, S.; Ertl, S.; Halkola, A.

doi:10.1051/0004-6361/202244534

Cited by 5 publications

(1 citation statement)

References 124 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Over the recent years, deep-learning methods have proven extremely successful at accurately modeling strong lensing systems (Hezaveh et al 2017;Perreault-Levasseur et al 2017;Morningstar et al 2018;Coogan et al 2020;Legin et al 2021;Park et al 2021;Wagner-Carena et al 2021;Anau Montel et al 2022;Karchev et al 2022;Legin et al 2022;Mishra-Sharma & Yang 2022;Schmidt et al 2023;Schuldt et al 2023;Wagner-Carena et al 2023). More specifically, Morningstar et al (2019) demonstrated that recurrent convolutional neural networks can implicitly learn complex prior distributions from their training data to reconstruct pixelated undistorted images of strongly lensed sources successfully, circumventing the need to specify explicitly a prior distribution over those parameters.…”

Section: Introductionmentioning

confidence: 99%

Pixelated Reconstruction of Foreground Density and Background Surface Brightness in Gravitational Lensing Systems Using Recurrent Inference Machines

Adam

Perreault-Levasseur

Hezaveh

et al. 2023

ApJ

View full text Add to dashboard Cite

Modeling strong gravitational lenses in order to quantify distortions in the images of background sources and to reconstruct the mass density in foreground lenses has been a difficult computational challenge. As the quality of gravitational lens images increases, the task of fully exploiting the information they contain becomes computationally and algorithmically more difficult. In this work, we use a neural network based on the recurrent inference machine to reconstruct simultaneously an undistorted image of the background source and the lens mass density distribution as pixelated maps. The method iteratively reconstructs the model parameters (the image of the source and a pixelated density map) by learning the process of optimizing the likelihood given the data using the physical model (a ray-tracing simulation), regularized by a prior implicitly learned by the neural network through its training data. When compared to more traditional parametric models, the proposed method is significantly more expressive and can reconstruct complex mass distributions, which we demonstrate by using realistic lensing galaxies taken from the IllustrisTNG cosmological hydrodynamic simulation.

show abstract

Section: Introductionmentioning

confidence: 99%

Pixelated Reconstruction of Foreground Density and Background Surface Brightness in Gravitational Lensing Systems Using Recurrent Inference Machines

Adam

Perreault-Levasseur

Hezaveh

et al. 2023

ApJ

View full text Add to dashboard Cite

show abstract

Machine Learning Investigations for LSST: Strong Lens Mass Modeling and Photometric Redshift Estimation

Schuldt

2023

Astrophysics and Space Science Proceedings

View full text Add to dashboard Cite

Holismokes

Cañameras,

Schuldt,

Shu

et al. 2024

A&A

View full text Add to dashboard Cite

While supervised neural networks have become state of the art for identifying the rare strong gravitational lenses from large imaging data sets, their selection remains significantly affected by the large number and diversity of non-lens contaminants. This work evaluates and compares systematically the performance of neural networks in order to move towards a rapid selection of galaxy-scale strong lenses with minimal human input in the era of deep, wide-scale surveys. We used multiband images from PDR2 of the Hyper-Suprime Cam (HSC) Wide survey to build test sets mimicking an actual classification experiment, with 189 securely-identified strong lenses from the literature over the HSC footprint and 70\,910 non-lens galaxies in COSMOS covering representative lens-like morphologies. Multiple networks were trained on different sets of realistic strong-lens simulations and non-lens galaxies, with various architectures and data preprocessing, mainly using the deepest $gri$-bands. Most networks reached excellent area under the Receiver Operating Characteristic (ROC) curves on the test set of 71,099 objects, and we determined the ingredients to optimize the true positive rate for a total number of false positives equal to zero or 10 $ and TPR$_ $). The overall performances strongly depend on the construction of the ground-truth training data and they typically, but not systematically, improve using our baseline residual network architecture presented in HOLISMOKES VI. TPR$_ $ tends to be higher for ResNets (simeq 10--40<!PCT!>) compared to AlexNet-like networks or G-CNNs. Improvements are found when (1) applying random shifts to the image centroids, (2) using square-root scaled images to enhance faint arcs, (3) adding $z$-band to the otherwise used $gri$-bands, or (4) using random viewpoints of the original images. In contrast, we find no improvement when adding $g - i$ difference images (where alpha is a tuned constant) to subtract emission from the central galaxy. The most significant gain is obtained with committees of networks trained on different data sets, with a moderate overlap between populations of false positives. Nearly-perfect invariance to image quality can be achieved by using realistic PSF models in our lens simulation pipeline, and by training networks either with large number of bands, or jointly with the PSF and science frames. Overall, we show the possibility to reach a $ as high as 60<!PCT!> for the test sets under consideration, which opens promising perspectives for pure selection of strong lenses without human input using the Rubin Observatory and other forthcoming ground-based surveys.

show abstract

Holismokes

Cited by 5 publications

References 124 publications

Pixelated Reconstruction of Foreground Density and Background Surface Brightness in Gravitational Lensing Systems Using Recurrent Inference Machines

Pixelated Reconstruction of Foreground Density and Background Surface Brightness in Gravitational Lensing Systems Using Recurrent Inference Machines

Machine Learning Investigations for LSST: Strong Lens Mass Modeling and Photometric Redshift Estimation

Holismokes

Contact Info

Product

Resources

About