Detecting gravitational lenses using machine learning: exploring interpretability and sensitivity to rare lensing configurations

TL;DR

This paper develops and evaluates CNN models for detecting strong gravitational lenses in large astronomical surveys, demonstrating high accuracy, sensitivity to rare lens types, and exploring interpretability methods to understand model decisions.

Contribution

It introduces CNN-based methods for gravitational lens detection that are effective on simulated and real data, and pioneers interpretability analysis for these models.

Findings

CNNs achieve F1 scores between 0.83 and 0.91.

CNNs do not bias against rare compound lenses, with high recall.

Interpretability techniques reveal how CNNs identify lens features.

Abstract

Forthcoming large imaging surveys such as Euclid and the Vera Rubin Observatory Legacy Survey of Space and Time are expected to find more than $10^5$ strong gravitational lens systems, including many rare and exotic populations such as compound lenses, but these $10^5$ systems will be interspersed among much larger catalogues of $\sim10^9$ galaxies. This volume of data is too much for visual inspection by volunteers alone to be feasible and gravitational lenses will only appear in a small fraction of these data which could cause a large amount of false positives. Machine learning is the obvious alternative but the algorithms' internal workings are not obviously interpretable, so their selection functions are opaque and it is not clear whether they would select against important rare populations. We design, build, and train several Convolutional Neural Networks (CNNs) to identify strong gravitational lenses using VIS, Y, J, and H bands of simulated data, with F1 scores between 0.83 and 0.91 on 100,000 test set images. We demonstrate for the first time that such CNNs do not select against compound lenses, obtaining recall scores as high as 76\% for compound arcs and 52\% for double rings. We verify this performance using Hubble Space Telescope (HST) and Hyper Suprime-Cam (HSC) data of all known compound lens systems. Finally, we explore for the first time the interpretability of these CNNs using Deep Dream, Guided Grad-CAM, and by exploring the kernels of the convolutional layers, to illuminate why CNNs succeed in compound lens selection.