Detecting gravitational lenses using machine learning: exploring interpretability and sensitivity to rare lensing configurations
Published in MNRAS, 2022
Recommended citation: Joshua Wilde, Stephen Serjeant, Jane M Bromley, Hugh Dickinson, Léon V E Koopmans, R Benton Metcalf, "Detecting gravitational lenses using machine learning: exploring interpretability and sensitivity to rare lensing configurations ", Monthly Notices of the Royal Astronomical Society, 2022;, stac562, https://doi.org/10.1093/mnras/stac562 http://academicpages.github.io/files/paper1.pdf
This paper is about using convolutional neural networks (CNNs) to identify strong gravitational lenses in simulated data. The CNN is applied to simulated compound gravitational lenses, and applied to HST imaging of four known compound gravitational lenses. The trained CNN is investigated to attempt to gain an understanding of what features the CNN has learnt. The GitHub for this paper can be found here LensFindery-McLensFinderFace
Recommended citation: Joshua Wilde, Stephen Serjeant, Jane M Bromley, Hugh Dickinson, Léon V E Koopmans, R Benton Metcalf, Detecting gravitational lenses using machine learning: exploring interpretability and sensitivity to rare lensing configurations, Monthly Notices of the Royal Astronomical Society, 2022;, stac562, https://doi.org/10.1093/mnras/stac562