Journal Article

DeepZipper: A Novel Deep-learning Architecture for Lensed Supernovae Identification

Annis, J, Park, J. W, To, C, Frieman, J, Palmese, A, García-Bellido, J, Miquel, R, González, S. J, Maia, M. A. G, Gaztanaga, E, Tarle, G, Nord, B, Gruen, D, Kuehn, K, Bechtol, K, Buckley-Geer, E, Plazas Malagón, A. A, Suchyta, E, Honscheid, K, Sanchez, E, Möller, A, Carretero, J, Kim, A. G, Paz-Chinchón, F, Swanson, M. E. C, Cawthon, R, Gruendl, R. A, Hollowood, D. L, Friedel, D, De Vicente, J, Brooks, D, da Costa, L. N, Doel, P, Birrer, S, Pereira, M. E. S, Giannini, G, Carnero Rosell, A, Smith, M, James, D. J, Ferrero, I, Roodman, A, Carrasco Kind, M, Gutierrez, G, Bocquet, S, Kuropatkin, N, Morgan, R, Reil, K, Aguena, M, Pieres, A, Davis, T. M, Gatti, M

2022

Overview

Large-scale astronomical surveys have the potential to capture data on large numbers of strongly gravitationally lensed supernovae (LSNe). To facilitate timely analysis and spectroscopic follow-up before the supernova fades, an LSN needs to be identified soon after it begins. To quickly identify LSNe in optical survey data sets, we designed ZipperNet, a multibranch deep neural network that combines convolutional layers (traditionally used for images) with long short-term memory layers (traditionally used for time series). We tested ZipperNet on the task of classifying objects from four categories—no lens, galaxy-galaxy lens, lensed Type-Ia supernova, lensed core-collapse supernova—within high-fidelity simulations of three cosmic survey data sets: the Dark Energy Survey, Rubin Observatory’s Legacy Survey of Space and Time (LSST), and a Dark Energy Spectroscopic Instrument (DESI) imaging survey. Among our results, we find that for the LSST-like data set, ZipperNet classifies LSNe with a receiver operating characteristic area under the curve of 0.97, predicts the spectroscopic type of the lensed supernovae with 79% accuracy, and demonstrates similarly high performance for LSNe 1–2 epochs after first detection. We anticipate that a model like ZipperNet, which simultaneously incorporates spatial and temporal information, can play a significant role in the rapid identification of lensed transient systems in cosmic survey experiments.

• Publisher: IOP Publishing
• Subject: Artificial neural networks / Astronomy / Celestial bodies / Dark energy / Datasets / Deep learning / Galaxies / Gravitational lenses / Long short-term memory / Machine learning / Neural networks / Polls & surveys / Sky surveys (astronomy) / Supernova / Supernovae