J/other/RAA/22.Fivo://CDS.VizieR/J/other/RAA/22.F5014doi:10.26093/cds/vizier/dx/cckqfabeCDSZhong F.Li R.Napolitano N.R.2026-02-10T20:08:00Z2023-08-30T13:42:47ZCDS support team

CDS, Observatoire de Strasbourg, 11 rue de l'Universite, F-67000 Strasbourg, France

cds-question@unistra.frgravitational-lensingredshiftedvisible-astronomyWith 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 techniques. To prepare for this challenge, we introduce a family of deep learning tools to classify features in one-dimensional spectra. As the first application of these Galaxy Spectra neural Networks (GaSNets), we focus on tools specialized in identifying emission lines from strongly lensed star-forming galaxies in the eBOSS spectra. We first discuss the training and testing of these networks and define a threshold probability, PL, of 95% for the high-quality event detection. Then, using a previous set of spectroscopically selected strong lenses from eBOSS, confirmed with the Hubble Space Telescope (HST), we estimate a completeness of ~80% as the fraction of lenses recovered above the adopted PL. We finally apply the GaSNets 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 HST. This 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.2022RAA....22f5014Zhttps://cdsarc.cds.unistra.fr/viz-bin/cat/J/other/RAA/22.F5014CatalogResearchIsServedByTAP VizieR generic serviceIsServedByConesearch servicerelated-toV/154 : Sloan Digital Sky Surveys (SDSS), Release 16 (DR16) (Ahumada+, 2020)https://cds.unistra.fr/vizier-org/licences_vizier.htmlhttps://vizier.cds.unistra.fr/viz-bin/VizieR-2?-source=J/other/RAA/22.F5014https://vizier.iucaa.in/viz-bin/VizieR-2?-source=J/other/RAA/22.F5014http://vizieridia.saao.ac.za/viz-bin/VizieR-2?-source=J/other/RAA/22.F5014https://vizier.cds.unistra.fr/viz-bin/votable?-source=J/other/RAA/22.F5014https://vizier.iucaa.in/viz-bin/votable?-source=J/other/RAA/22.F5014http://vizieridia.saao.ac.za/viz-bin/votable?-source=J/other/RAA/22.F5014GETtext/xml+votablehttps://tapvizier.cds.unistra.fr/TAPVizieR/tapCone search capability for table J/other/RAA/22.F5014/appena (New high quality (HQ) candidates from GaSNets)https://vizier.cds.unistra.fr/viz-bin/conesearch/J/other/RAA/22.F5014/appena?https://vizier.iucaa.in/viz-bin/conesearch/J/other/RAA/22.F5014/appena?http://vizieridia.saao.ac.za/viz-bin/conesearch/J/other/RAA/22.F5014/appena?GETtext/xml+votable180.050000true27.6755-7.16670.0055555555555555566/653 690 696 2057 2086 2097 2107 2127 2140 2157-2158 2190 2202 2212 2232 2307 2561 4097 4107 4155 4229 4234 4238 4319 4348 4397 4410 4466 4533 4536 4541 4584 4643 4671 4693 4698 4710 4736 4739 4771 4792 4846 4912 4967 4971 4979 5034 5075 5077 5093 5107 5163 5180 5203 5233 5332 5362 5399 5475 5499 5516 5536 5550 5558 5562 5568 5589 5629 5640 5679 5699 5704 5711 5733 5795 5804 5807 5811 5849 5883 5914 5920 5931 5934 5965 5978 5985 5996 6054 6092 6101 6162 6173 6224 6353 6405 6457 6481 6503 6629 7244 7256 7262 7332-7333 7395 7406 7458 7488 7497 7512 7520 7555 8248 8286 8338 8406 8424 8431 8446 8454-8455 8500 8511 8533 8635 8687 8749 8780 8810 8867 8874 8879 9052 9131 9195 9275 9290 9302 9316 9397 9426 9429 9516 9521 9600 9607 9623 9789 9795 9887 9896 9926 10029 10038 10040 10120 10122 10285 10358 10380 10439 10458 10460 10538 10596 10641 10643 10701 10717 10746 10895 10919 10951 10978 11006 11022 11027 11044 11087 11145 11160 11167 11179 11235 11256 11333 11340 11348 11369 11419 11494 11504 11544 11565 11792 11834 11875 12316 12352 12618 12758 12761 13327 13351 13374 13411 13415 13424 13426 13477 13489 13510 13586 17276 17303 17529 17585 17603 17650 17709 17722 17729 17798 17812-17813 17819 17858 17890 18002 18054 18063 18077 18106 18117 18131 18157 18187 18191 18217 18236 18254-18255 18312 18351 18395 18405 18421 18685 18757 18834 19017 19060 19090 19097 19105 19121 19126 19143 19172 19212 19229 19237 19244 19250 19270 19305 19329 19419 19426 19474 19510 19539 19546 19549 19615 19675 19697 19702 19771 19791 19797 19832 19836-19837 19925 19945 19952 20034 20060 20088 20097 20193 20212 20290 20301 20313 20339 20364 20388-20389 20424 20451 21960 22387 25847 25915 25950 25992 25996 26039 26042 26073 26086 26163 26234 26261 26316 26327 26344 26347 26358 26388 26393 26407 26441-26442 26450 26469 26492 26534 26547 26588 26592 26601 26611 26971 27050 27087 27259 27314 27336 27365 27409-27410 27486 27490 27502 27528 27545 27564 27601 27629 27666 27684 27701 27760 27762 27822 27830-27831 27835-27836 27848 27853 27910 27964 28008 28106 28129 28135 28166 28185-28186 28224 28246 28264 28269 28277 28371 28426 28429 28463 28571 28575 28595 28630 28635 28643 28649 28660 31379 31435 31438 31449 31456 31465 31627 31686 31735 36778-36779 36784 36793 36799 36847 45043https://cdsarc.cds.unistra.fr/viz-bin/moc/J/other/RAA/22.F5014?format=asciiOpticaldefaultJ/other/RAA/22.F5014/appenaNew high quality (HQ) candidates from GaSNetsSimbadNameSimbad column added by the CDSmeta.idcharnullablerecnoRecord number assigned by the VizieR team. Should Not be used for identification.meta.recordintSeqSequential numbermeta.recordintRAJ2000Right ascension (J2000)degpos.eq.ra;meta.mainDEJ2000Declination (J2000)degpos.eq.dec;meta.mainPlatePlate identifiermeta.id;instr.plateintMJDMJD identifiertime.epochintFiberFiber identifiermeta.id;meta.mainintPLGaSNet-L1 probabilitystat.fit.goodnessfloatzRedshift of the galaxy from the eBOSS catalogsrc.redshiftfloatzPGPredicted redshift of the galaxy from GaSNet-L3src.redshift;pos.heliocentricfloatzPEPredicted redshift of the background source from GaSNet-L2src.redshift;pos.heliocentricfloatmeaneyeZECorrected redshift of the source from the visual inspectionsrc.redshift;pos.heliocentricfloatPTTotal probability, PT=PLx0.1 visual scores, i.e., combining the GaSNets and human probabilities to be a lensstat.fit.goodnessfloatSpInteractive spectrum from SDSSmeta.ref.urlchar