Home > Research > Publications & Outputs > DeepZipper. II. Searching for Lensed Supernovae...

Research

Links

Text available via DOI:

View graph of relations

DeepZipper. II. Searching for Lensed Supernovae in Dark Energy Survey Data with Deep Learning

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

DeepZipper. II. Searching for Lensed Supernovae in Dark Energy Survey Data with Deep Learning. / Dark Energy Survey Collaboration.
In: The Astrophysical Journal, Vol. 943, No. 1, 19, 20.01.2023.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Dark Energy Survey Collaboration 2023, 'DeepZipper. II. Searching for Lensed Supernovae in Dark Energy Survey Data with Deep Learning', The Astrophysical Journal, vol. 943, no. 1, 19. https://doi.org/10.3847/1538-4357/ac721b

APA

Dark Energy Survey Collaboration (2023). DeepZipper. II. Searching for Lensed Supernovae in Dark Energy Survey Data with Deep Learning. The Astrophysical Journal, 943(1), Article 19. https://doi.org/10.3847/1538-4357/ac721b

Vancouver

Dark Energy Survey Collaboration. DeepZipper. II. Searching for Lensed Supernovae in Dark Energy Survey Data with Deep Learning. The Astrophysical Journal. 2023 Jan 20;943(1):19. doi: 10.3847/1538-4357/ac721b

Author

Dark Energy Survey Collaboration. / DeepZipper. II. Searching for Lensed Supernovae in Dark Energy Survey Data with Deep Learning. In: The Astrophysical Journal. 2023 ; Vol. 943, No. 1.

Bibtex

@article{3780f94ae41947f9acbf302f32bd8159,
title = "DeepZipper. II. Searching for Lensed Supernovae in Dark Energy Survey Data with Deep Learning",
abstract = "Gravitationally lensed supernovae (LSNe) are important probes of cosmic expansion, but they remain rare and difficult to find. Current cosmic surveys likely contain 5-10 LSNe in total while next-generation experiments are expected to contain several hundred to a few thousand of these systems. We search for these systems in observed Dark Energy Survey (DES) five year SN fields—10 3 sq. deg. regions of sky imaged in the griz bands approximately every six nights over five years. To perform the search, we utilize the DeepZipper approach: a multi-branch deep learning architecture trained on image-level simulations of LSNe that simultaneously learns spatial and temporal relationships from time series of images. We find that our method obtains an LSN recall of 61.13% and a false-positive rate of 0.02% on the DES SN field data. DeepZipper selected 2245 candidates from a magnitude-limited (m i < 22.5) catalog of 3,459,186 systems. We employ human visual inspection to review systems selected by the network and find three candidate LSNe in the DES SN fields.",
author = "{Dark Energy Survey Collaboration} and R. Morgan and B. Nord and K. Bechtol and A. M{\"o}ller and Hartley, {W. G.} and S. Birrer and Gonz{\'a}lez, {S. J.} and M. Martinez and Gruendl, {R. A.} and Buckley-Geer, {E. J.} and Shajib, {A. J.} and {Carnero Rosell}, A. and C. Lidman and T. Collett and Abbott, {T. M. C.} and M. Aguena and F. Andrade-Oliveira and J. Annis and D. Bacon and S. Bocquet and D. Brooks and Burke, {D. L.} and {Carrasco Kind}, M. and J. Carretero and Castander, {F. J.} and C. Conselice and {da Costa}, {L. N.} and M. Costanzi and {De Vicente}, J. and S. Desai and P. Doel and S. Everett and I. Ferrero and B. Flaugher and D. Friedel and J. Frieman and J. Garc{\'i}a-Bellido and E. Gaztanaga and D. Gruen and G. Gutierrez and Hinton, {S. R.} and Hollowood, {D. L.} and K. Honscheid and K. Kuehn and N. Kuropatkin and O. Lahav and M. Lima and F. Menanteau and R. Miquel and M. Smith",
year = "2023",
month = jan,
day = "20",
doi = "10.3847/1538-4357/ac721b",
language = "English",
volume = "943",
journal = "The Astrophysical Journal",
issn = "0004-637X",
publisher = "Institute of Physics Publishing",
number = "1",

}

RIS

TY - JOUR

T1 - DeepZipper. II. Searching for Lensed Supernovae in Dark Energy Survey Data with Deep Learning

AU - Dark Energy Survey Collaboration

AU - Morgan, R.

AU - Nord, B.

AU - Bechtol, K.

AU - Möller, A.

AU - Hartley, W. G.

AU - Birrer, S.

AU - González, S. J.

AU - Martinez, M.

AU - Gruendl, R. A.

AU - Buckley-Geer, E. J.

AU - Shajib, A. J.

AU - Carnero Rosell, A.

AU - Lidman, C.

AU - Collett, T.

AU - Abbott, T. M. C.

AU - Aguena, M.

AU - Andrade-Oliveira, F.

AU - Annis, J.

AU - Bacon, D.

AU - Bocquet, S.

AU - Brooks, D.

AU - Burke, D. L.

AU - Carrasco Kind, M.

AU - Carretero, J.

AU - Castander, F. J.

AU - Conselice, C.

AU - da Costa, L. N.

AU - Costanzi, M.

AU - De Vicente, J.

AU - Desai, S.

AU - Doel, P.

AU - Everett, S.

AU - Ferrero, I.

AU - Flaugher, B.

AU - Friedel, D.

AU - Frieman, J.

AU - García-Bellido, J.

AU - Gaztanaga, E.

AU - Gruen, D.

AU - Gutierrez, G.

AU - Hinton, S. R.

AU - Hollowood, D. L.

AU - Honscheid, K.

AU - Kuehn, K.

AU - Kuropatkin, N.

AU - Lahav, O.

AU - Lima, M.

AU - Menanteau, F.

AU - Miquel, R.

AU - Smith, M.

PY - 2023/1/20

Y1 - 2023/1/20

N2 - Gravitationally lensed supernovae (LSNe) are important probes of cosmic expansion, but they remain rare and difficult to find. Current cosmic surveys likely contain 5-10 LSNe in total while next-generation experiments are expected to contain several hundred to a few thousand of these systems. We search for these systems in observed Dark Energy Survey (DES) five year SN fields—10 3 sq. deg. regions of sky imaged in the griz bands approximately every six nights over five years. To perform the search, we utilize the DeepZipper approach: a multi-branch deep learning architecture trained on image-level simulations of LSNe that simultaneously learns spatial and temporal relationships from time series of images. We find that our method obtains an LSN recall of 61.13% and a false-positive rate of 0.02% on the DES SN field data. DeepZipper selected 2245 candidates from a magnitude-limited (m i < 22.5) catalog of 3,459,186 systems. We employ human visual inspection to review systems selected by the network and find three candidate LSNe in the DES SN fields.

AB - Gravitationally lensed supernovae (LSNe) are important probes of cosmic expansion, but they remain rare and difficult to find. Current cosmic surveys likely contain 5-10 LSNe in total while next-generation experiments are expected to contain several hundred to a few thousand of these systems. We search for these systems in observed Dark Energy Survey (DES) five year SN fields—10 3 sq. deg. regions of sky imaged in the griz bands approximately every six nights over five years. To perform the search, we utilize the DeepZipper approach: a multi-branch deep learning architecture trained on image-level simulations of LSNe that simultaneously learns spatial and temporal relationships from time series of images. We find that our method obtains an LSN recall of 61.13% and a false-positive rate of 0.02% on the DES SN field data. DeepZipper selected 2245 candidates from a magnitude-limited (m i < 22.5) catalog of 3,459,186 systems. We employ human visual inspection to review systems selected by the network and find three candidate LSNe in the DES SN fields.

U2 - 10.3847/1538-4357/ac721b

DO - 10.3847/1538-4357/ac721b

M3 - Journal article

VL - 943

JO - The Astrophysical Journal

JF - The Astrophysical Journal

SN - 0004-637X

IS - 1

M1 - 19

ER -