SNIascore: Deep-learning Classification of Low-resolution Supernova Spectra
SNIascore: Deep-learning Classification of Low-resolution Supernova Spectra
复制标题
SNIascore：低分辨率超新星光谱的深度学习分类
DOI：
--
复制
发表时间：
2021
影响因子：
7.9
通讯作者：
S. Kulkarni
中科院分区：
文献类型：
--
作者：
C. Fremling;X. Hall;M. Coughlin;A. Dahiwale;D. Duev;M. Graham;M. Kasliwal;E. Kool;Adam A. Miller;J. Neill;D. Perley;M. Rigault;P. Rosnet;B. Rusholme;Y. Sharma;K. Shin;D. Shupe;J. Sollerman;R. Walters;S. Kulkarni
我们提出了SNIascore，这是一种基于深度学习的方法，用于基于极低分辨率（R = 100）数据的热核超新星（SNe Ia）光谱分类。SNIascore的目标是以非常低的假阳性率（FPR）对SNe Ia进行全自动分类，以便在大规模SN分类工作中大大减少人为干预，例如由公共Zwicky Transient Facility（ZTF）Bright Transient Survey（BTS）进行的分类工作。我们利用一个循环神经网络架构，结合双向长短期记忆和门控循环单元层。SNIascore实现了<0.6%的FPR，同时对BTS获得的高达90%的低分辨率SN Ia光谱进行分类。SNIascore同时执行二进制分类，并通过回归预测安全Ia超新星的红移（在z = 0.01至z = 0.12的范围内，典型的不确定性<0.005）。对于星等受限的ZTF BTS勘测（SNe Ia = 1.70%），部署SNIascore可将需要人工分类或确认的光谱数量减少1.60%。此外，SNIascore还允许在夜间完成观测后立即以真实的时间向公众自动宣布超新星Ia的分类。
We present SNIascore, a deep-learning-based method for spectroscopic classification of thermonuclear supernovae (SNe Ia) based on very low-resolution (R ∼ 100) data. The goal of SNIascore is the fully automated classification of SNe Ia with a very low false-positive rate (FPR) so that human intervention can be greatly reduced in large-scale SN classification efforts, such as that undertaken by the public Zwicky Transient Facility (ZTF) Bright Transient Survey (BTS). We utilize a recurrent neural network architecture with a combination of bidirectional long short-term memory and gated recurrent unit layers. SNIascore achieves a <0.6% FPR while classifying up to 90% of the low-resolution SN Ia spectra obtained by the BTS. SNIascore simultaneously performs binary classification and predicts the redshifts of secure SNe Ia via regression (with a typical uncertainty of <0.005 in the range from z = 0.01 to z = 0.12). For the magnitude-limited ZTF BTS survey (≈70% SNe Ia), deploying SNIascore reduces the amount of spectra in need of human classification or confirmation by ≈60%. Furthermore, SNIascore allows SN Ia classifications to be automatically announced in real time to the public immediately following a finished observation during the night.
DOI：
10.1088/1538-3873/aaa53f
发表时间：
2018-03-01
影响因子：
3.5
作者：
Blagorodnova, Nadejda;Neill, James D.;Vyhmeister, Karl E.
通讯作者：
Vyhmeister, Karl E.