Galaxy Spectra Neural Networks(GaSNets). I. Searching for Strong Lens Candidates in eBOSS Spectra Using Deep Learning

作     者：Fucheng Zhong Rui Li Nicola R.Napolitano Fucheng Zhong;Rui Li;Nicola R.Napolitano

作者机构：School of Physics and AstronomySun Yat-sen UniversityZhuhai CampusZhuhai 519082China CSST Science Center for Guangdong-Hong Kong-Macao Great Bay AreaZhuhai 519082China School of Astronomy and Space ScienceUniversity of Chinese Academy of SciencesBeijing 100049China National Astronomical ObservatoriesChinese Academy of SciencesBeijing 100101China 

出 版 物：《Research in Astronomy and Astrophysics》 (天文和天体物理学研究（英文版）)

年 卷 期：2022年第22卷第6期

页      面：142-169页

核心收录：

学科分类：07[理学] 070401[理学-天体物理] 0704[理学-天文学] 

基　　金：the science research grants from the China Manned Space Project(CMSCSST-2021-A01) the support from K.C.Wong Education Foundation financial support from the“One hundred top talent program of Sun Yatsen University”grant No.71000-18841229 from the Research Fund for International Scholars of the National Science Foundation of China,grant No.12150710511 

主　　题：gravitational lensing strong-surveys-techniques spectroscopic 

摘      要：With the advent of new spectroscopic surveys from ground and space,observing up to hundreds of millions of galaxies,spectra classification will become overwhelming for standard analysis *** prepare for this challenge,we introduce a family of deep learning tools to classify features in one-dimensional *** the first application of these Galaxy Spectra neural Networks(Ga SNets),we focus on tools specialized in identifying emission lines from strongly lensed star-forming galaxies in the e BOSS *** first discuss the training and testing of these networks and define a threshold probability,PL,of 95%for the high-quality event ***,using a previous set of spectroscopically selected strong lenses from e BOSS,confirmed with the Hubble Space Telescope(HST),we estimate a completeness of~80%as the fraction of lenses recovered above the adopted *** finally apply the Ga SNets to~1.3M eBOSS spectra to collect the first list of~430 new high-quality candidates identified with deep learning from spectroscopy and visually graded as highly probable real events.A preliminary check against ground-based observations tentatively shows that this sample has a confirmation rate of 38%,in line with previous samples selected with standard(no deep learning)classification tools and confirmed by the *** first test shows that machine learning can be efficiently extended to feature recognition in the wavelength space,which will be crucial for future surveys like 4MOST,DESI,Euclid,and the China Space Station Telescope.