DeepZipper: A Novel Deep Learning Architecture for Lensed Supernovae Identification

Autor: Morgan, Robert, Nord, B., Bechtol, K., González, S. J., Buckley-Geer, E., Möller, A., Park, J. W., Kim, A. G., Birrer, S., Aguena, M., Annis, J., Bocquet, S., Brooks, D., Rosell, A. Carnero, Kind, M. Carrasco, Carretero, J., Cawthon, R., da Costa, L. N., Davis, T. M., De Vicente, J., Doel, P., Ferrero, I., Friedel, D., Frieman, J., García-Bellido, J., Gatti, M., Gaztanaga, E., Giannini, G., Gruen, D., Gruendl, R. A., Gutierrez, G., Hollowood, D. L., Honscheid, K., James, D. J., Kuehn, K., Kuropatkin, N., Maia, M. A. G., Miquel, R., Palmese, A., Paz-Chinchón, F., Pereira, M. E. S., Pieres, A., Malagón, A. A. Plazas, Reil, K., Roodman, A., Sanchez, E., Smith, M., Suchyta, E., Swanson, M. E. C., Tarle, G., To, C.
Rok vydání: 2021
Předmět:
Druh dokumentu: Working Paper
DOI: 10.3847/1538-4357/ac5178
Popis: 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 datasets, we designed ZipperNet, a multi-branch deep neural network that combines convolutional layers (traditionally used for images) with long short-term memory (LSTM) 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 (DES), 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 dataset, 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.
Comment: Published in ApJ
Databáze: arXiv
Externí odkaz: