Cargando…

Automated Detection of Classical Novae with Neural Networks

The POINT-AGAPE collaboration surveyed M31 with the primary goal of optical detection of microlensing events, yet its data catalogue is also a prime source of lightcurves of variable and transient objects, including classical novae (CNe). A reliable means of identification, combined with a thorough...

Descripción completa

Detalles Bibliográficos
Autores principales: Feeney, S., Belokurov, V., Evans, N.W., An, J., Hewett, Paul C., Bode, M., Darnley, M., Kerins, E., Baillon, P., Carr, Bernard J., Paulin-Henriksson, S., Gould, A.
Lenguaje:eng
Publicado: 2005
Materias:
Acceso en línea:https://dx.doi.org/10.1086/430844
http://cds.cern.ch/record/832142
_version_ 1780905829911756800
author Feeney, S.
Belokurov, V.
Evans, N.W.
An, J.
Hewett, Paul C.
Bode, M.
Darnley, M.
Kerins, E.
Baillon, P.
Carr, Bernard J.
Paulin-Henriksson, S.
Gould, A.
author_facet Feeney, S.
Belokurov, V.
Evans, N.W.
An, J.
Hewett, Paul C.
Bode, M.
Darnley, M.
Kerins, E.
Baillon, P.
Carr, Bernard J.
Paulin-Henriksson, S.
Gould, A.
author_sort Feeney, S.
collection CERN
description The POINT-AGAPE collaboration surveyed M31 with the primary goal of optical detection of microlensing events, yet its data catalogue is also a prime source of lightcurves of variable and transient objects, including classical novae (CNe). A reliable means of identification, combined with a thorough survey of the variable objects in M31, provides an excellent opportunity to locate and study an entire galactic population of CNe. This paper presents a set of 440 neural networks, working in 44 committees, designed specifically to identify fast CNe. The networks are developed using training sets consisting of simulated novae and POINT-AGAPE lightcurves, in a novel variation on K-fold cross-validation. They use the binned, normalised power spectra of the lightcurves as input units. The networks successfully identify 9 of the 13 previously identified M31 CNe within their optimal working range (and 11 out of 13 if the network error bars are taken into account). They provide a catalogue of 19 new candidate fast CNe, of which 4 are strongly favoured.
id cern-832142
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2005
record_format invenio
spelling cern-8321422023-03-14T19:53:44Zdoi:10.1086/430844http://cds.cern.ch/record/832142engFeeney, S.Belokurov, V.Evans, N.W.An, J.Hewett, Paul C.Bode, M.Darnley, M.Kerins, E.Baillon, P.Carr, Bernard J.Paulin-Henriksson, S.Gould, A.Automated Detection of Classical Novae with Neural NetworksAstrophysics and AstronomyThe POINT-AGAPE collaboration surveyed M31 with the primary goal of optical detection of microlensing events, yet its data catalogue is also a prime source of lightcurves of variable and transient objects, including classical novae (CNe). A reliable means of identification, combined with a thorough survey of the variable objects in M31, provides an excellent opportunity to locate and study an entire galactic population of CNe. This paper presents a set of 440 neural networks, working in 44 committees, designed specifically to identify fast CNe. The networks are developed using training sets consisting of simulated novae and POINT-AGAPE lightcurves, in a novel variation on K-fold cross-validation. They use the binned, normalised power spectra of the lightcurves as input units. The networks successfully identify 9 of the 13 previously identified M31 CNe within their optimal working range (and 11 out of 13 if the network error bars are taken into account). They provide a catalogue of 19 new candidate fast CNe, of which 4 are strongly favoured.The POINT-AGAPE collaboration surveyed M31 with the primary goal of optical detection of microlensing events, yet its data catalogue is also a prime source of lightcurves of variable and transient objects, including classical novae (CNe). A reliable means of identification, combined with a thorough survey of the variable objects in M31, provides an excellent opportunity to locate and study an entire galactic population of CNe. This paper presents a set of 440 neural networks, working in 44 committees, designed specifically to identify fast CNe. The networks are developed using training sets consisting of simulated novae and POINT-AGAPE lightcurves, in a novel variation on K-fold cross-validation. They use the binned, normalised power spectra of the lightcurves as input units. The networks successfully identify 9 of the 13 previously identified M31 CNe within their optimal working range (and 11 out of 13 if the network error bars are taken into account). They provide a catalogue of 19 new candidate fast CNe, of which 4 are strongly favoured.astro-ph/0504236oai:cds.cern.ch:8321422005-04-11
spellingShingle Astrophysics and Astronomy
Feeney, S.
Belokurov, V.
Evans, N.W.
An, J.
Hewett, Paul C.
Bode, M.
Darnley, M.
Kerins, E.
Baillon, P.
Carr, Bernard J.
Paulin-Henriksson, S.
Gould, A.
Automated Detection of Classical Novae with Neural Networks
title Automated Detection of Classical Novae with Neural Networks
title_full Automated Detection of Classical Novae with Neural Networks
title_fullStr Automated Detection of Classical Novae with Neural Networks
title_full_unstemmed Automated Detection of Classical Novae with Neural Networks
title_short Automated Detection of Classical Novae with Neural Networks
title_sort automated detection of classical novae with neural networks
topic Astrophysics and Astronomy
url https://dx.doi.org/10.1086/430844
http://cds.cern.ch/record/832142
work_keys_str_mv AT feeneys automateddetectionofclassicalnovaewithneuralnetworks
AT belokurovv automateddetectionofclassicalnovaewithneuralnetworks
AT evansnw automateddetectionofclassicalnovaewithneuralnetworks
AT anj automateddetectionofclassicalnovaewithneuralnetworks
AT hewettpaulc automateddetectionofclassicalnovaewithneuralnetworks
AT bodem automateddetectionofclassicalnovaewithneuralnetworks
AT darnleym automateddetectionofclassicalnovaewithneuralnetworks
AT kerinse automateddetectionofclassicalnovaewithneuralnetworks
AT baillonp automateddetectionofclassicalnovaewithneuralnetworks
AT carrbernardj automateddetectionofclassicalnovaewithneuralnetworks
AT paulinhenrikssons automateddetectionofclassicalnovaewithneuralnetworks
AT goulda automateddetectionofclassicalnovaewithneuralnetworks