PyTorchDIA: a flexible, GPU-accelerated numerical approach to Difference Image Analysis

We present a GPU-accelerated numerical approach for fast kernel and differential background solutions. The model image proposed in the Bramich Difference Image Analysis (DIA) algorithm is analogous to a very simple convolutional neural network (CNN), with a single convolutional filter (i.e. the kern...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Hitchcock, James (VerfasserIn) , Hundertmark, Markus (VerfasserIn) , Foreman-Mackey, Daniel (VerfasserIn) , Bachelet, Etienne (VerfasserIn) , Dominik, Martin (VerfasserIn) , Street, Rachel (VerfasserIn) , Tsapras, Yiannis (VerfasserIn)
Dokumenttyp: Article (Journal)
Sprache:Englisch
Veröffentlicht: 21 April 2021
In: Monthly notices of the Royal Astronomical Society
Year: 2021, Jahrgang: 504, Heft: 3, Pages: 3561-3579
ISSN:1365-2966
DOI:10.1093/mnras/stab1114
Online-Zugang:Verlag, lizenzpflichtig, Volltext: https://doi.org/10.1093/mnras/stab1114
Volltext
Verfasserangaben:James A. Hitchcock, Markus Hundertmark, Daniel Foreman-Mackey, Etienne Bachelet, Martin Dominik, Rachel Street and Yiannis Tsapras

MARC

LEADER 00000caa a2200000 c 4500
001 1765412838
003 DE-627
005 20220820030659.0
007 cr uuu---uuuuu
008 210804s2021 xx |||||o 00| ||eng c
024 7 |a 10.1093/mnras/stab1114  |2 doi 
035 |a (DE-627)1765412838 
035 |a (DE-599)KXP1765412838 
035 |a (OCoLC)1341419984 
040 |a DE-627  |b ger  |c DE-627  |e rda 
041 |a eng 
084 |a 29  |2 sdnb 
100 1 |a Hitchcock, James  |e VerfasserIn  |0 (DE-588)1219930717  |0 (DE-627)1736028197  |4 aut 
245 1 0 |a PyTorchDIA  |b a flexible, GPU-accelerated numerical approach to Difference Image Analysis  |c James A. Hitchcock, Markus Hundertmark, Daniel Foreman-Mackey, Etienne Bachelet, Martin Dominik, Rachel Street and Yiannis Tsapras 
264 1 |c 21 April 2021 
300 |a 19 
336 |a Text  |b txt  |2 rdacontent 
337 |a Computermedien  |b c  |2 rdamedia 
338 |a Online-Ressource  |b cr  |2 rdacarrier 
500 |a Gesehen am 04.08.2021 
520 |a We present a GPU-accelerated numerical approach for fast kernel and differential background solutions. The model image proposed in the Bramich Difference Image Analysis (DIA) algorithm is analogous to a very simple convolutional neural network (CNN), with a single convolutional filter (i.e. the kernel) and an added scalar bias (i.e. the differential background). Here, we do not solve for the discrete pixel array in the classical, analytical linear least-squares sense. Instead, by making use of PyTorch tensors (GPU compatible multidimensional matrices) and associated deep learning tools, we solve for the kernel via an inherently massively parallel optimization. By casting the DIA problem as a GPU-accelerated optimization that utilizes automatic differentiation tools, our algorithm is both flexible to the choice of scalar objective function, and can perform DIA on astronomical data sets at least an order of magnitude faster than its classical analogue. More generally, we demonstrate that tools developed for machine learning can be used to address generic data analysis and modelling problems. 
700 1 |a Hundertmark, Markus  |e VerfasserIn  |0 (DE-588)1138794627  |0 (DE-627)896371042  |0 (DE-576)492681937  |4 aut 
700 1 |a Foreman-Mackey, Daniel  |e VerfasserIn  |4 aut 
700 1 |a Bachelet, Etienne  |e VerfasserIn  |4 aut 
700 1 |a Dominik, Martin  |e VerfasserIn  |4 aut 
700 1 |a Street, Rachel  |e VerfasserIn  |4 aut 
700 1 |a Tsapras, Yiannis  |d 1975-  |e VerfasserIn  |0 (DE-588)1093855932  |0 (DE-627)853882037  |0 (DE-576)407996400  |4 aut 
773 0 8 |i Enthalten in  |a Royal Astronomical Society  |t Monthly notices of the Royal Astronomical Society  |d Oxford : Oxford Univ. Press, 1827  |g 504(2021), 3 vom: Juli, Seite 3561-3579  |h Online-Ressource  |w (DE-627)314059164  |w (DE-600)2016084-7  |w (DE-576)090955420  |x 1365-2966  |7 nnas 
773 1 8 |g volume:504  |g year:2021  |g number:3  |g month:07  |g pages:3561-3579  |g extent:19  |a PyTorchDIA a flexible, GPU-accelerated numerical approach to Difference Image Analysis 
856 4 0 |u https://doi.org/10.1093/mnras/stab1114  |x Verlag  |x Resolving-System  |z lizenzpflichtig  |3 Volltext 
951 |a AR 
992 |a 20210804 
993 |a Article 
994 |a 2021 
998 |g 1093855932  |a Tsapras, Yiannis  |m 1093855932:Tsapras, Yiannis  |d 700000  |d 714000  |d 714100  |e 700000PT1093855932  |e 714000PT1093855932  |e 714100PT1093855932  |k 0/700000/  |k 1/700000/714000/  |k 2/700000/714000/714100/  |p 7  |y j 
998 |g 1138794627  |a Hundertmark, Markus  |m 1138794627:Hundertmark, Markus  |d 700000  |d 714000  |d 714100  |e 700000PH1138794627  |e 714000PH1138794627  |e 714100PH1138794627  |k 0/700000/  |k 1/700000/714000/  |k 2/700000/714000/714100/  |p 2 
999 |a KXP-PPN1765412838  |e 3961551960 
BIB |a Y 
SER |a journal 
JSO |a {"id":{"doi":["10.1093/mnras/stab1114"],"eki":["1765412838"]},"note":["Gesehen am 04.08.2021"],"language":["eng"],"recId":"1765412838","name":{"displayForm":["James A. Hitchcock, Markus Hundertmark, Daniel Foreman-Mackey, Etienne Bachelet, Martin Dominik, Rachel Street and Yiannis Tsapras"]},"relHost":[{"type":{"bibl":"periodical","media":"Online-Ressource"},"disp":"Royal Astronomical SocietyMonthly notices of the Royal Astronomical Society","physDesc":[{"extent":"Online-Ressource"}],"note":["Gesehen am 15.01.2018"],"id":{"zdb":["2016084-7"],"eki":["314059164"],"doi":["10.1111/(ISSN)1365-2966"],"issn":["1365-2966"]},"recId":"314059164","origin":[{"dateIssuedDisp":"1827-","publisher":"Oxford Univ. Press ; Blackwell ; Wiley-Blackwell","publisherPlace":"Oxford ; Oxford [u.a.] ; Oxford [u.a.]","dateIssuedKey":"1827"}],"corporate":[{"display":"Royal Astronomical Society","roleDisplay":"VerfasserIn","role":"aut"}],"pubHistory":["1.1827 -"],"title":[{"title":"Monthly notices of the Royal Astronomical Society","title_sort":"Monthly notices of the Royal Astronomical Society"}],"part":{"volume":"504","year":"2021","extent":"19","issue":"3","text":"504(2021), 3 vom: Juli, Seite 3561-3579","pages":"3561-3579"},"language":["eng"]}],"origin":[{"dateIssuedKey":"2021","dateIssuedDisp":"21 April 2021"}],"type":{"media":"Online-Ressource","bibl":"article-journal"},"title":[{"subtitle":"a flexible, GPU-accelerated numerical approach to Difference Image Analysis","title":"PyTorchDIA","title_sort":"PyTorchDIA"}],"person":[{"roleDisplay":"VerfasserIn","given":"James","role":"aut","family":"Hitchcock","display":"Hitchcock, James"},{"family":"Hundertmark","display":"Hundertmark, Markus","given":"Markus","roleDisplay":"VerfasserIn","role":"aut"},{"role":"aut","given":"Daniel","roleDisplay":"VerfasserIn","family":"Foreman-Mackey","display":"Foreman-Mackey, Daniel"},{"display":"Bachelet, Etienne","family":"Bachelet","given":"Etienne","roleDisplay":"VerfasserIn","role":"aut"},{"display":"Dominik, Martin","family":"Dominik","role":"aut","roleDisplay":"VerfasserIn","given":"Martin"},{"role":"aut","roleDisplay":"VerfasserIn","given":"Rachel","family":"Street","display":"Street, Rachel"},{"roleDisplay":"VerfasserIn","given":"Yiannis","role":"aut","family":"Tsapras","display":"Tsapras, Yiannis"}],"physDesc":[{"extent":"19 S."}]} 
SRT |a HITCHCOCKJPYTORCHDIA2120