Home > Research > Publications & Outputs > CCSNscore

Research

Associated organisational unit

Links

Text available via DOI:

View graph of relations

CCSNscore: A Multi-input Deep Learning Tool for Classification of Core-collapse Supernovae Using SED-machine Spectra

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

Published

Standard

CCSNscore: A Multi-input Deep Learning Tool for Classification of Core-collapse Supernovae Using SED-machine Spectra. / Sharma, Y.; Mahabal, A.A.; Sollerman, J. et al.
In: Publications of the Astronomical Society of the Pacific, Vol. 137, No. 3, 28.03.2025.

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

Harvard

Sharma, Y, Mahabal, AA, Sollerman, J, Fremling, C, Kulkarni, SR, Rehemtulla, N, Miller, AA, Aubert, M, Chen, TX, Coughlin, MW, Graham, MJ, Hale, D, Kasliwal, MM, Kim, Y-L, Neill, JD, Purdum, JN, Rusholme, B, Singh, A & Sravan, N 2025, 'CCSNscore: A Multi-input Deep Learning Tool for Classification of Core-collapse Supernovae Using SED-machine Spectra', Publications of the Astronomical Society of the Pacific, vol. 137, no. 3. https://doi.org/10.1088/1538-3873/adbf4b

APA

Sharma, Y., Mahabal, A. A., Sollerman, J., Fremling, C., Kulkarni, S. R., Rehemtulla, N., Miller, A. A., Aubert, M., Chen, T. X., Coughlin, M. W., Graham, M. J., Hale, D., Kasliwal, M. M., Kim, Y.-L., Neill, J. D., Purdum, J. N., Rusholme, B., Singh, A., & Sravan, N. (2025). CCSNscore: A Multi-input Deep Learning Tool for Classification of Core-collapse Supernovae Using SED-machine Spectra. Publications of the Astronomical Society of the Pacific, 137(3). https://doi.org/10.1088/1538-3873/adbf4b

Vancouver

Sharma Y, Mahabal AA, Sollerman J, Fremling C, Kulkarni SR, Rehemtulla N et al. CCSNscore: A Multi-input Deep Learning Tool for Classification of Core-collapse Supernovae Using SED-machine Spectra. Publications of the Astronomical Society of the Pacific. 2025 Mar 28;137(3). doi: 10.1088/1538-3873/adbf4b

Author

Sharma, Y. ; Mahabal, A.A. ; Sollerman, J. et al. / CCSNscore : A Multi-input Deep Learning Tool for Classification of Core-collapse Supernovae Using SED-machine Spectra. In: Publications of the Astronomical Society of the Pacific. 2025 ; Vol. 137, No. 3.

Bibtex

@article{2f3aed8df85242c0bbc1fb86831bba70,
title = "CCSNscore: A Multi-input Deep Learning Tool for Classification of Core-collapse Supernovae Using SED-machine Spectra",
abstract = "Supernovae (SNe) come in various flavors and are classified into different types based on emission and absorption lines in their spectra. SN candidates are now abundant with the advent of large systematic sky surveys like the Zwicky Transient Facility (ZTF), however, the identification bottleneck lies in their spectroscopic confirmation and classification. Fully robotic telescopes with dedicated spectrographs optimized for SN follow-up have eased the burden of data acquisition. However, the task of classifying the spectra still largely rests with the astronomers. Automating this classification step reduces human effort and can make the SN type available sooner to the public. For this purpose, we have developed a deep-learning based program for classifying core-collapse supernovae (CCSNe) with ultra-low resolution spectra from the SED-machine spectrograph on the Palomar 60 inch telescope. The program consists of hierarchical classification task layers, with each layer composed of multiple binary classifiers running in parallel to produce a reliable classification. The binary classifiers utilize recurrent neural networks and convolutional neural networks architecture and are designed to take multiple inputs to supplement spectra with g- and r-band photometry from ZTF. On non-host-contaminated and good quality SEDM spectra ({"}gold{"} test set) of CCSNe, CCSNscore is ∼94% accurate in distinguishing between hydrogen-rich (Type II) and hydrogen-poor (Type Ibc) CCSNe. With light curve input, CCSNscore classifies ∼83% of the gold set with high confidence (score ≥0.8 and score-error < 0.05), with ∼98% accuracy. Based on SNIascore's and CCSNscore's real-time performance on bright transients (mpk ≤ 18.5) and our reporting criteria, we expect ∼0.5% (∼4) true SNe Ia to be misclassified as SNe Ibc and ∼6% (∼17) of true CCSNe to be misclassified between Type II and Type Ibc annually on the Transient Name Server.",
author = "Y. Sharma and A.A. Mahabal and J. Sollerman and C. Fremling and S.R. Kulkarni and N. Rehemtulla and A.A. Miller and M. Aubert and T.X. Chen and M.W. Coughlin and M.J. Graham and D. Hale and M.M. Kasliwal and Y.-L. Kim and J.D. Neill and J.N. Purdum and B. Rusholme and A. Singh and N. Sravan",
year = "2025",
month = mar,
day = "28",
doi = "10.1088/1538-3873/adbf4b",
language = "English",
volume = "137",
journal = "Publications of the Astronomical Society of the Pacific",
issn = "0004-6280",
publisher = "University of Chicago",
number = "3",

}

RIS

TY - JOUR

T1 - CCSNscore

T2 - A Multi-input Deep Learning Tool for Classification of Core-collapse Supernovae Using SED-machine Spectra

AU - Sharma, Y.

AU - Mahabal, A.A.

AU - Sollerman, J.

AU - Fremling, C.

AU - Kulkarni, S.R.

AU - Rehemtulla, N.

AU - Miller, A.A.

AU - Aubert, M.

AU - Chen, T.X.

AU - Coughlin, M.W.

AU - Graham, M.J.

AU - Hale, D.

AU - Kasliwal, M.M.

AU - Kim, Y.-L.

AU - Neill, J.D.

AU - Purdum, J.N.

AU - Rusholme, B.

AU - Singh, A.

AU - Sravan, N.

PY - 2025/3/28

Y1 - 2025/3/28

N2 - Supernovae (SNe) come in various flavors and are classified into different types based on emission and absorption lines in their spectra. SN candidates are now abundant with the advent of large systematic sky surveys like the Zwicky Transient Facility (ZTF), however, the identification bottleneck lies in their spectroscopic confirmation and classification. Fully robotic telescopes with dedicated spectrographs optimized for SN follow-up have eased the burden of data acquisition. However, the task of classifying the spectra still largely rests with the astronomers. Automating this classification step reduces human effort and can make the SN type available sooner to the public. For this purpose, we have developed a deep-learning based program for classifying core-collapse supernovae (CCSNe) with ultra-low resolution spectra from the SED-machine spectrograph on the Palomar 60 inch telescope. The program consists of hierarchical classification task layers, with each layer composed of multiple binary classifiers running in parallel to produce a reliable classification. The binary classifiers utilize recurrent neural networks and convolutional neural networks architecture and are designed to take multiple inputs to supplement spectra with g- and r-band photometry from ZTF. On non-host-contaminated and good quality SEDM spectra ("gold" test set) of CCSNe, CCSNscore is ∼94% accurate in distinguishing between hydrogen-rich (Type II) and hydrogen-poor (Type Ibc) CCSNe. With light curve input, CCSNscore classifies ∼83% of the gold set with high confidence (score ≥0.8 and score-error < 0.05), with ∼98% accuracy. Based on SNIascore's and CCSNscore's real-time performance on bright transients (mpk ≤ 18.5) and our reporting criteria, we expect ∼0.5% (∼4) true SNe Ia to be misclassified as SNe Ibc and ∼6% (∼17) of true CCSNe to be misclassified between Type II and Type Ibc annually on the Transient Name Server.

AB - Supernovae (SNe) come in various flavors and are classified into different types based on emission and absorption lines in their spectra. SN candidates are now abundant with the advent of large systematic sky surveys like the Zwicky Transient Facility (ZTF), however, the identification bottleneck lies in their spectroscopic confirmation and classification. Fully robotic telescopes with dedicated spectrographs optimized for SN follow-up have eased the burden of data acquisition. However, the task of classifying the spectra still largely rests with the astronomers. Automating this classification step reduces human effort and can make the SN type available sooner to the public. For this purpose, we have developed a deep-learning based program for classifying core-collapse supernovae (CCSNe) with ultra-low resolution spectra from the SED-machine spectrograph on the Palomar 60 inch telescope. The program consists of hierarchical classification task layers, with each layer composed of multiple binary classifiers running in parallel to produce a reliable classification. The binary classifiers utilize recurrent neural networks and convolutional neural networks architecture and are designed to take multiple inputs to supplement spectra with g- and r-band photometry from ZTF. On non-host-contaminated and good quality SEDM spectra ("gold" test set) of CCSNe, CCSNscore is ∼94% accurate in distinguishing between hydrogen-rich (Type II) and hydrogen-poor (Type Ibc) CCSNe. With light curve input, CCSNscore classifies ∼83% of the gold set with high confidence (score ≥0.8 and score-error < 0.05), with ∼98% accuracy. Based on SNIascore's and CCSNscore's real-time performance on bright transients (mpk ≤ 18.5) and our reporting criteria, we expect ∼0.5% (∼4) true SNe Ia to be misclassified as SNe Ibc and ∼6% (∼17) of true CCSNe to be misclassified between Type II and Type Ibc annually on the Transient Name Server.

U2 - 10.1088/1538-3873/adbf4b

DO - 10.1088/1538-3873/adbf4b

M3 - Journal article

VL - 137

JO - Publications of the Astronomical Society of the Pacific

JF - Publications of the Astronomical Society of the Pacific

SN - 0004-6280

IS - 3

ER -