DeepMerge – II. Building robust deep learning algorithms for merging galaxy identification across domains

In astronomy, neural networks are often trained on simulation data with the prospect of being used on telescope observations. Unfortunately, training a model on simulation data and then applying it to instrument data leads to a substantial and potentially even detrimental decrease in model accuracy...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Monthly notices of the Royal Astronomical Society Ročník 506; číslo 1; s. 677 - 691
Hlavní autoři:	Ćiprijanović, A, Kafkes, D, Downey, K, Jenkins, S, Perdue, G N, Madireddy, S, Johnston, T, Snyder, G F, Nord, B
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	United Kingdom Oxford University Press 01.09.2021
Témata:	ASTRONOMY AND ASTROPHYSICS galaxies: evolution galaxies: interactions INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND TECHNOLOGY methods: data analysis techniques: image processing
ISSN:	0035-8711, 1365-2966
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Abstract	In astronomy, neural networks are often trained on simulation data with the prospect of being used on telescope observations. Unfortunately, training a model on simulation data and then applying it to instrument data leads to a substantial and potentially even detrimental decrease in model accuracy on the new target data set. Simulated and instrument data represent different data domains, and for an algorithm to work in both, domain-invariant learning is necessary. Here, we employ domain adaptation techniques – Maximum Mean Discrepancy as an additional transfer loss and Domain Adversarial Neural Networks – and demonstrate their viability to extract domain-invariant features within the astronomical context of classifying merging and non-merging galaxies. Additionally, we explore the use of Fisher loss and entropy minimization to enforce better in-domain class discriminability. We show that the addition of each domain adaptation technique improves the performance of a classifier when compared to conventional deep learning algorithms. We demonstrate this on two examples: between two Illustris-1 simulated data sets of distant merging galaxies, and between Illustris-1 simulated data of nearby merging galaxies and observed data from the Sloan Digital Sky Survey. The use of domain adaptation techniques in our experiments leads to an increase of target domain classification accuracy of up to ${\sim }20{{\ \rm per\ cent}}$. With further development, these techniques will allow astronomers to successfully implement neural network models trained on simulation data to efficiently detect and study astrophysical objects in current and future large-scale astronomical surveys.
AbstractList	In astronomy, neural networks are often trained on simulation data with the prospect of being used on telescope observations. Unfortunately, training a model on simulation data and then applying it to instrument data leads to a substantial and potentially even detrimental decrease in model accuracy on the new target data set. Simulated and instrument data represent different data domains, and for an algorithm to work in both, domain-invariant learning is necessary. Here, we employ domain adaptation techniques – Maximum Mean Discrepancy as an additional transfer loss and Domain Adversarial Neural Networks – and demonstrate their viability to extract domain-invariant features within the astronomical context of classifying merging and non-merging galaxies. Additionally, we explore the use of Fisher loss and entropy minimization to enforce better in-domain class discriminability. We show that the addition of each domain adaptation technique improves the performance of a classifier when compared to conventional deep learning algorithms. We demonstrate this on two examples: between two Illustris-1 simulated data sets of distant merging galaxies, and between Illustris-1 simulated data of nearby merging galaxies and observed data from the Sloan Digital Sky Survey. The use of domain adaptation techniques in our experiments leads to an increase of target domain classification accuracy of up to ${\sim }20{{\ \rm per\ cent}}$. With further development, these techniques will allow astronomers to successfully implement neural network models trained on simulation data to efficiently detect and study astrophysical objects in current and future large-scale astronomical surveys. ABSTRACT In astronomy, neural networks are often trained on simulation data with the prospect of being used on telescope observations. Unfortunately, training a model on simulation data and then applying it to instrument data leads to a substantial and potentially even detrimental decrease in model accuracy on the new target data set. Simulated and instrument data represent different data domains, and for an algorithm to work in both, domain-invariant learning is necessary. Here, we employ domain adaptation techniques – Maximum Mean Discrepancy as an additional transfer loss and Domain Adversarial Neural Networks – and demonstrate their viability to extract domain-invariant features within the astronomical context of classifying merging and non-merging galaxies. Additionally, we explore the use of Fisher loss and entropy minimization to enforce better in-domain class discriminability. We show that the addition of each domain adaptation technique improves the performance of a classifier when compared to conventional deep learning algorithms. We demonstrate this on two examples: between two Illustris-1 simulated data sets of distant merging galaxies, and between Illustris-1 simulated data of nearby merging galaxies and observed data from the Sloan Digital Sky Survey. The use of domain adaptation techniques in our experiments leads to an increase of target domain classification accuracy of up to ${\sim }20{{\ \rm per\ cent}}$. With further development, these techniques will allow astronomers to successfully implement neural network models trained on simulation data to efficiently detect and study astrophysical objects in current and future large-scale astronomical surveys.
Author	Downey, K Kafkes, D Madireddy, S Perdue, G N Johnston, T Snyder, G F Ćiprijanović, A Jenkins, S Nord, B
Author_xml	– sequence: 1 givenname: A orcidid: 0000-0003-1281-7192 surname: Ćiprijanović fullname: Ćiprijanović, A – sequence: 2 givenname: D surname: Kafkes fullname: Kafkes, D – sequence: 3 givenname: K surname: Downey fullname: Downey, K – sequence: 4 givenname: S surname: Jenkins fullname: Jenkins, S – sequence: 5 givenname: G N surname: Perdue fullname: Perdue, G N – sequence: 6 givenname: S surname: Madireddy fullname: Madireddy, S – sequence: 7 givenname: T surname: Johnston fullname: Johnston, T – sequence: 8 givenname: G F orcidid: 0000-0002-4226-304X surname: Snyder fullname: Snyder, G F – sequence: 9 givenname: B orcidid: 0000-0001-6706-8972 surname: Nord fullname: Nord, B
BackLink	https://www.osti.gov/biblio/1807592$$D View this record in Osti.gov
BookMark	eNp1kL1OwzAUhS1UJEphZbbY0_oncZIRyl-lIhaYoxvbSQ2JXdmuRDfegTfkSUhbWJCYrnR0vqNzzykaWWc1QheUTCkp-ay3HsIsRKipyPMjNKZcZAkrhRihMSE8S4qc0hN0GsIrISTlTIzR243W60ftW42_Pj7xYjHF1xvTKWNb7F29CRGrwYE7Dd7uROha501c9QE3zuN-QHdyCx28b7FR2kbTGAnROItBehcCVq4HY8MZOm6gC_r8507Qy93t8_whWT7dL-ZXy0RyzmLCicoVU7zUGRGCqLSAPCso15LpomS0YXVDQAoCqpR1nvJUFpRlQFUOTGTAJ-jykOtCNFWQJmq5ks5aLWNFC5JnJRtM04NpX9Hrplp704PfVpRUuz2r_Z7V754DkP4BhuD9m9GD6f7DvgGhpYGG
CitedBy_id	crossref_primary_10_1051_0004_6361_202348188 crossref_primary_10_3847_1538_4365_ad66ca crossref_primary_10_1093_mnras_stac1609 crossref_primary_10_1051_0004_6361_202346743 crossref_primary_10_1017_pasa_2022_55 crossref_primary_10_1093_mnras_stad414 crossref_primary_10_3847_1538_4365_add0b8 crossref_primary_10_1093_mnras_stad1895 crossref_primary_10_1093_mnras_stac1241 crossref_primary_10_3847_1538_4365_acab02 crossref_primary_10_1088_2632_2153_acca5f crossref_primary_10_1093_mnras_stad3181 crossref_primary_10_3847_1538_4357_ac8c27 crossref_primary_10_3847_1538_4357_adee16 crossref_primary_10_3847_1538_4357_acdfc7 crossref_primary_10_1051_0004_6361_202452799 crossref_primary_10_1088_2632_2153_ac98f4 crossref_primary_10_3847_2041_8213_ad8dd4 crossref_primary_10_1051_0004_6361_202452659 crossref_primary_10_1155_2022_6758684 crossref_primary_10_3847_1538_3881_ad1661 crossref_primary_10_1088_2632_2153_ac7f1a crossref_primary_10_3847_1538_4357_ac66ea crossref_primary_10_1051_epjconf_202226500009 crossref_primary_10_1093_mnras_stab3717 crossref_primary_10_1093_mnras_stac3295
Cites_doi	10.1214/aoms/1177729694 10.1093/mnras/stz1059 10.1007/978-3-319-58347-1_1 10.1093/mnras/stx2656 10.1109/CVPR.2019.00503 10.1093/mnras/stv2078 10.3847/1538-4357/aabfed 10.1086/377318 10.1093/mnras/sty1398 10.1111/j.1365-2966.2010.17432.x 10.1146/annurev-astro-081811-125615 10.1137/1.9781611976236.53 10.1111/j.1468-4004.2009.50512.x 10.23915/distill.00002 10.1093/mnras/stx3040 10.1145/3400066 10.1109/TNN.2010.2091281 10.1111/j.1365-2966.2008.13689.x 10.1007/978-3-319-49409-8_35 10.1086/427183 10.1109/CVPR.2009.5206848 10.1093/mnras/stx3304 10.1016/j.neucom.2018.05.083 10.1093/mnras/stu2058 10.1093/mnras/stu1536 10.1145/3295500.3356202 10.1111/j.1365-2966.2009.15686.x 10.1016/j.ascom.2020.100390 10.1086/421849 10.1093/mnras/staa2777 10.1111/j.1365-2966.2006.11287.x 10.1109/CVPR.2016.90 10.1093/mnras/stz2934 10.1093/mnras/staa3778 10.1111/j.1365-2966.2007.11458.x 10.1051/0004-6361/202038084 10.3847/1538-4357/ab8f9b 10.1051/0004-6361/202037963 10.1109/HiPC.2018.00014 10.1002/j.1538-7305.1948.tb01338.x 10.1051/0004-6361/201935355 10.1007/s11263-019-01228-7 10.1109/TPAMI.2016.2599532 10.1186/s40668-019-0028-x
ContentType	Journal Article
CorporateAuthor	Fermi National Accelerator Laboratory (FNAL), Batavia, IL (United States)Argonne National Laboratory (ANL), Argonne, IL (United States). Argonne Leadership Computing Facility (ALCF)
CorporateAuthor_xml	– name: Fermi National Accelerator Laboratory (FNAL), Batavia, IL (United States)Argonne National Laboratory (ANL), Argonne, IL (United States). Argonne Leadership Computing Facility (ALCF)
DBID	AAYXX CITATION OTOTI
DOI	10.1093/mnras/stab1677
DatabaseName	CrossRef OSTI.GOV
DatabaseTitle	CrossRef
DatabaseTitleList	CrossRef
DeliveryMethod	fulltext_linktorsrc
Discipline	Meteorology & Climatology Astronomy & Astrophysics
EISSN	1365-2966
EndPage	691
ExternalDocumentID	1807592 10_1093_mnras_stab1677
GroupedDBID	-DZ -~X .2P .I3 0R~ 123 1OC 29M 4.4 48X 51W 51X 52X 5HH 5LA 5VS 66C 6TJ 702 7PT 8UM AAHTB AAIJN AAJKP AAKDD AAMVS AAOGV AAPQZ AAPXW AARHZ AAVAP AAYXX ABCQN ABCQX ABEJV ABEML ABEUO ABGNP ABIXL ABNKS ABPEJ ABPTD ABQLI ABVLG ABXVV ABZBJ ACGFO ACGFS ACGOD ACNCT ACSCC ACUFI ACUXJ ACYRX ACYTK ADEYI ADGZP ADHKW ADHZD ADOCK ADQBN ADRDM ADRTK ADVEK ADYVW ADZXQ AECKG AEGPL AEJOX AEKKA AEKSI AEMDU AENEX AENZO AEPUE AETBJ AEWNT AFEBI AFFZL AFIYH AFOFC AGINJ AGSYK AHGBF AHXPO AJEEA AJEUX ALMA_UNASSIGNED_HOLDINGS ALTZX ALUQC ALXQX AMNDL ANAKG APIBT AVWKF AXUDD AZVOD BAYMD BEFXN BEYMZ BFFAM BGNUA BHONS BKEBE BPEOZ BQUQU BTQHN CDBKE CITATION CO8 DAKXR DILTD DR2 DU5 D~K E.L E3Z EBS EE~ F5P F9B FLIZI FLUFQ FOEOM FRJ GAUVT GJXCC GROUPED_DOAJ H13 H5~ HAR HF~ HW0 HZI HZ~ IHE IX1 J21 JAVBF JXSIZ K48 KBUDW KOP KQ8 KSI KSN L7B LC2 LC3 LH4 LP6 LP7 M43 MK4 NGC NMDNZ NOMLY O9- OCL ODMLO OIG OJQWA OK1 P2P P2X P4D PAFKI PEELM PQQKQ Q1. Q11 Q5Y ROX ROZ RUSNO RW1 RX1 RXO TJP TN5 TOX V8K W8V WH7 X5Q X5S YAYTL YKOAZ YXANX 69O AABJS AABMN AAESY AAIYJ AAPBV ABPTK ADEIU ADIPN ADORX ADQLU ADRIX AFUVZ AFXEN AIKOY ARQIP AUCZF AZQFJ BCRHZ BY8 BYORX CASEJ DPORF DPPUQ ESX OTOTI OWPYF PQEST RHF UMP WRC
ID	FETCH-LOGICAL-c332t-30d7d2d39e50660d48a75813ec2e8921f2bf0ac60ad9cb7434c8125a1d7a265a3
ISICitedReferencesCount	29
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000703918800044&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	0035-8711
IngestDate	Fri May 19 00:43:38 EDT 2023 Tue Nov 18 20:48:37 EST 2025 Sat Nov 29 02:38:22 EST 2025
IsPeerReviewed	true
IsScholarly	true
Issue	1
Language	English
License	https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model
LinkModel	OpenURL
MergedId	FETCHMERGED-LOGICAL-c332t-30d7d2d39e50660d48a75813ec2e8921f2bf0ac60ad9cb7434c8125a1d7a265a3
Notes	AC02-06CH11357; AC02-07CH11359 USDOE Office of Science (SC), High Energy Physics (HEP) FERMILAB-PUB-21-072-SCD; arXiv:2103.01373
ORCID	0000-0003-1281-7192 0000-0001-6706-8972 0000-0002-4226-304X 0000000312817192 000000024226304X 0000000167068972
PageCount	15
ParticipantIDs	osti_scitechconnect_1807592 crossref_primary_10_1093_mnras_stab1677 crossref_citationtrail_10_1093_mnras_stab1677
PublicationCentury	2000
PublicationDate	2021-09-01
PublicationDateYYYYMMDD	2021-09-01
PublicationDate_xml	– month: 09 year: 2021 text: 2021-09-01 day: 01
PublicationDecade	2020
PublicationPlace	United Kingdom
PublicationPlace_xml	– name: United Kingdom
PublicationTitle	Monthly notices of the Royal Astronomical Society
PublicationYear	2021
Publisher	Oxford University Press
Publisher_xml	– name: Oxford University Press
References	Ganin (2022061100144889200_bib15) 2016; 17 Kullback (2022061100144889200_bib27) 1951; 22 Balaprakash (2022061100144889200_bib2) 2018 Sun (2022061100144889200_bib51) 2016 Shannon (2022061100144889200_bib43) 1948; 27 Venkateswara (2022061100144889200_bib53) 2017 van der Maaten (2022061100144889200_bib52) 2008; 9 Gretton (2022061100144889200_bib19) 2012 Shen (2022061100144889200_bib44) 2018 Kitzbichler (2022061100144889200_bib25) 2007; 376 Ćiprijanović (2022061100144889200_bib61) 2020; 32 Huertas-Company (2022061100144889200_bib22) 2018; 858 Pearson (2022061100144889200_bib37) 2019; 626 Driver (2022061100144889200_bib13) 2009; 50 Wattenberg (2022061100144889200_bib57) 2016 Zhang (2022061100144889200_bib59) 2020 Gretton (2022061100144889200_bib20) 2012; 13 He (2022061100144889200_bib21) 2016 Balaprakash (2022061100144889200_bib3) 2019 Darg (2022061100144889200_bib9) 2010; 401 Madau (2022061100144889200_bib33) 2014; 52 Lin (2022061100144889200_bib28) 2004; 617 Nelson (2022061100144889200_bib34) 2018; 475 Ackermann (2022061100144889200_bib1) 2018; 479 Ginzburg (2022061100144889200_bib17) 2021; 501 Snyder (2022061100144889200_bib47) 2015; 454 Wang (2022061100144889200_bib56) 2020; 644 Grandvalet (2022061100144889200_bib18) 2005 Wilson (2022061100144889200_bib58) 2020; 11 Chollet (2022061100144889200_bib6) 2016 Saenko (2022061100144889200_bib39) 2010 Ghifary (2022061100144889200_bib16) 2017; 39 Pillepich (2022061100144889200_bib38) 2018; 473 Liu (2022061100144889200_bib31) 2016 Simonyan (2022061100144889200_bib45) 2015 Pan (2022061100144889200_bib36) 2011; 22 Vogelsberger (2022061100144889200_bib54) 2014; 444 Cavanagh (2022061100144889200_bib5) 2020; 641 Lintott (2022061100144889200_bib30) 2010; 410 Huertas-Company (2022061100144889200_bib23) 2020; 499 Bottrell (2022061100144889200_bib4) 2019; 490 Deng (2022061100144889200_bib12) 2009 Schaye (2022061100144889200_bib40) 2015; 446 Koppula (2022061100144889200_bib26) 2021 Smola (2022061100144889200_bib46) 2007 De Lucia (2022061100144889200_bib11) 2007; 375 Sérsic (2022061100144889200_bib42) 1963; 6 Conselice (2022061100144889200_bib7) 2003; 126 Sun (2022061100144889200_bib50) 2016 Csurka (2022061100144889200_bib8) 2017 Lotz (2022061100144889200_bib32) 2004; 128 Springel (2022061100144889200_bib49) 2018; 475 Zhang (2022061100144889200_bib60) 2020 de Jong (2022061100144889200_bib10) 2013; 154 Lintott (2022061100144889200_bib29) 2008; 389 Nelson (2022061100144889200_bib35) 2019; 6 Snyder (2022061100144889200_bib48) 2019; 486 Wang (2022061100144889200_bib55) 2018; 312 Kang (2022061100144889200_bib24) 2019 Ferreira (2022061100144889200_bib14) 2020; 895 Selvaraju (2022061100144889200_bib41) 2020; 128
References_xml	– volume: 22 start-page: 79 year: 1951 ident: 2022061100144889200_bib27 publication-title: Ann. Math. Statist. doi: 10.1214/aoms/1177729694 – volume: 486 start-page: 3702 year: 2019 ident: 2022061100144889200_bib48 publication-title: MNRAS doi: 10.1093/mnras/stz1059 – start-page: 1 volume-title: A Comprehensive Survey on Domain Adaptation for Visual Applications year: 2017 ident: 2022061100144889200_bib8 doi: 10.1007/978-3-319-58347-1_1 – volume: 473 start-page: 4077 year: 2018 ident: 2022061100144889200_bib38 publication-title: MNRAS doi: 10.1093/mnras/stx2656 – start-page: 13 volume-title: Algorithmic Learning Theory, Lecture Notes in Computer Science 4754 year: 2007 ident: 2022061100144889200_bib46 – start-page: 4888 volume-title: in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) year: 2019 ident: 2022061100144889200_bib24 doi: 10.1109/CVPR.2019.00503 – volume: 454 start-page: 1886 year: 2015 ident: 2022061100144889200_bib47 publication-title: MNRAS doi: 10.1093/mnras/stv2078 – volume: 858 start-page: 114 year: 2018 ident: 2022061100144889200_bib22 publication-title: ApJ doi: 10.3847/1538-4357/aabfed – volume: 126 start-page: 1183 year: 2003 ident: 2022061100144889200_bib7 publication-title: AJ doi: 10.1086/377318 – volume: 154 start-page: 44 year: 2013 ident: 2022061100144889200_bib10 publication-title: Messenger – volume: 479 start-page: 415 year: 2018 ident: 2022061100144889200_bib1 publication-title: MNRAS doi: 10.1093/mnras/sty1398 – volume: 410 start-page: 166 year: 2010 ident: 2022061100144889200_bib30 publication-title: MNRAS doi: 10.1111/j.1365-2966.2010.17432.x – start-page: 469 volume-title: Advances in Neural Information Processing Systems 29: Annual Conference on Neural Information Processing Systems year: 2016 ident: 2022061100144889200_bib31 – start-page: 213 volume-title: Proceedings of the 11th European Conference on Computer Vision: Part IV. ECCV’10 year: 2010 ident: 2022061100144889200_bib39 – volume: 52 start-page: 415 year: 2014 ident: 2022061100144889200_bib33 publication-title: Ann. Rev. Astron. Astrophys. doi: 10.1146/annurev-astro-081811-125615 – start-page: 469 volume-title: Proceedings of the 2020 SIAM International Conference on Data Mining (SDM) year: 2020 ident: 2022061100144889200_bib59 doi: 10.1137/1.9781611976236.53 – volume: 17 start-page: 1 year: 2016 ident: 2022061100144889200_bib15 publication-title: J. Mach. Learn. Res. – volume: 50 start-page: 5.12 year: 2009 ident: 2022061100144889200_bib13 publication-title: A&G doi: 10.1111/j.1468-4004.2009.50512.x – volume: 6 start-page: 99 year: 1963 ident: 2022061100144889200_bib42 publication-title: Boletin de la Asociacion Argentina de Astronomia La Plata Argentina – year: 2016 ident: 2022061100144889200_bib57 article-title: Distill, How to Use t-SNE Effectively doi: 10.23915/distill.00002 – volume: 475 start-page: 624 year: 2018 ident: 2022061100144889200_bib34 publication-title: MNRAS doi: 10.1093/mnras/stx3040 – volume: 11 start-page: 1 year: 2020 ident: 2022061100144889200_bib58 publication-title: ACM Trans. Intell. Syst. Technol. doi: 10.1145/3400066 – year: 2016 ident: 2022061100144889200_bib6 publication-title: CoRR – volume: 22 start-page: 199 year: 2011 ident: 2022061100144889200_bib36 publication-title: Trans. Neur. Netw. doi: 10.1109/TNN.2010.2091281 – volume: 389 start-page: 1179 year: 2008 ident: 2022061100144889200_bib29 publication-title: MNRAS doi: 10.1111/j.1365-2966.2008.13689.x – start-page: 443 volume-title: Lecture Notes in Computer Science Vol. 9915, Computer Vision – ECCV 2016 Workshops – Amsterdam, Proceedings, Part III year: 2016 ident: 2022061100144889200_bib50 doi: 10.1007/978-3-319-49409-8_35 – volume: 617 start-page: L9 year: 2004 ident: 2022061100144889200_bib28 publication-title: ApJ doi: 10.1086/427183 – start-page: 248 volume-title: 2009 IEEE Conference on Computer Vision and Pattern Recognition year: 2009 ident: 2022061100144889200_bib12 doi: 10.1109/CVPR.2009.5206848 – start-page: 2058 volume-title: Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence. AAAI’16 year: 2016 ident: 2022061100144889200_bib51 – volume: 9 start-page: 2579 year: 2008 ident: 2022061100144889200_bib52 publication-title: J. Mach. Learn. Res. – volume: 475 start-page: 676 year: 2018 ident: 2022061100144889200_bib49 publication-title: MNRAS doi: 10.1093/mnras/stx3304 – start-page: 529 volume-title: Advances in Neural Information Processing Systems 17 year: 2005 ident: 2022061100144889200_bib18 – year: 2015 ident: 2022061100144889200_bib45 – volume: 312 start-page: 135 year: 2018 ident: 2022061100144889200_bib55 publication-title: Neurocomputing doi: 10.1016/j.neucom.2018.05.083 – volume: 446 start-page: 521 year: 2015 ident: 2022061100144889200_bib40 publication-title: MNRAS doi: 10.1093/mnras/stu2058 – volume: 444 start-page: 1518 year: 2014 ident: 2022061100144889200_bib54 publication-title: MNRAS doi: 10.1093/mnras/stu1536 – volume-title: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis. SC ’19 year: 2019 ident: 2022061100144889200_bib3 doi: 10.1145/3295500.3356202 – start-page: 1205 volume-title: Advances in Neural Information Processing Systems 25 year: 2012 ident: 2022061100144889200_bib19 – volume: 401 start-page: 1043 year: 2010 ident: 2022061100144889200_bib9 publication-title: MNRAS doi: 10.1111/j.1365-2966.2009.15686.x – volume: 32 start-page: 100390 year: 2020 ident: 2022061100144889200_bib61 publication-title: A&A doi: 10.1016/j.ascom.2020.100390 – volume: 128 start-page: 163 year: 2004 ident: 2022061100144889200_bib32 publication-title: AJ doi: 10.1086/421849 – volume: 499 start-page: 814 year: 2020 ident: 2022061100144889200_bib23 publication-title: MNRAS doi: 10.1093/mnras/staa2777 – volume: 375 start-page: 2 year: 2007 ident: 2022061100144889200_bib11 publication-title: MNRAS doi: 10.1111/j.1365-2966.2006.11287.x – start-page: 770 volume-title: in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) year: 2016 ident: 2022061100144889200_bib21 doi: 10.1109/CVPR.2016.90 – volume: 490 start-page: 5390 year: 2019 ident: 2022061100144889200_bib4 publication-title: MNRAS doi: 10.1093/mnras/stz2934 – volume: 501 start-page: 730 year: 2021 ident: 2022061100144889200_bib17 publication-title: MNRAS doi: 10.1093/mnras/staa3778 – volume: 376 start-page: 2 year: 2007 ident: 2022061100144889200_bib25 publication-title: MNRAS doi: 10.1111/j.1365-2966.2007.11458.x – volume: 644 start-page: A87 year: 2020 ident: 2022061100144889200_bib56 publication-title: A&A doi: 10.1051/0004-6361/202038084 – start-page: 4058 volume-title: Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, (AAAI-18), the 30th innovative Applications of Artificial Intelligence (IAAI-18), and the 8th AAAI Symposium on Educational Advances in Artificial Intelligence (EAAI-18) year: 2018 ident: 2022061100144889200_bib44 – volume: 895 start-page: 115 year: 2020 ident: 2022061100144889200_bib14 publication-title: ApJ doi: 10.3847/1538-4357/ab8f9b – volume: 641 start-page: A77 year: 2020 ident: 2022061100144889200_bib5 publication-title: A&A doi: 10.1051/0004-6361/202037963 – year: 2020 ident: 2022061100144889200_bib60 – start-page: 42 volume-title: 2018 IEEE 25th International Conference on High Performance Computing (HiPC) year: 2018 ident: 2022061100144889200_bib2 doi: 10.1109/HiPC.2018.00014 – volume: 27 start-page: 379 year: 1948 ident: 2022061100144889200_bib43 publication-title: Bell Syst. Tech. J. doi: 10.1002/j.1538-7305.1948.tb01338.x – volume: 626 start-page: A49 year: 2019 ident: 2022061100144889200_bib37 publication-title: A&A doi: 10.1051/0004-6361/201935355 – volume: 128 start-page: 336 year: 2020 ident: 2022061100144889200_bib41 publication-title: Int. J. Comput. Vis. doi: 10.1007/s11263-019-01228-7 – volume: 39 start-page: 1414 year: 2017 ident: 2022061100144889200_bib16 publication-title: IEEE TPAMI doi: 10.1109/TPAMI.2016.2599532 – year: 2017 ident: 2022061100144889200_bib53 publication-title: CoRR – volume: 13 start-page: 723 year: 2012 ident: 2022061100144889200_bib20 publication-title: J. Mach. Learn. Res. – year: 2021 ident: 2022061100144889200_bib26 – volume: 6 start-page: 2 year: 2019 ident: 2022061100144889200_bib35 publication-title: Comput. Astrophys. Cosmol. doi: 10.1186/s40668-019-0028-x
SSID	ssj0004326
Score	2.5972104
Snippet	In astronomy, neural networks are often trained on simulation data with the prospect of being used on telescope observations. Unfortunately, training a model... ABSTRACT In astronomy, neural networks are often trained on simulation data with the prospect of being used on telescope observations. Unfortunately, training...
SourceID	osti crossref
SourceType	Open Access Repository Enrichment Source Index Database
StartPage	677
SubjectTerms	ASTRONOMY AND ASTROPHYSICS galaxies: evolution galaxies: interactions INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND TECHNOLOGY methods: data analysis techniques: image processing
Title	DeepMerge – II. Building robust deep learning algorithms for merging galaxy identification across domains
URI	https://www.osti.gov/biblio/1807592
Volume	506
WOSCitedRecordID	wos000703918800044&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVASL databaseName: Oxford Journals Open Access Collection customDbUrl: eissn: 1365-2966 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0004326 issn: 0035-8711 databaseCode: TOX dateStart: 18591101 isFulltext: true titleUrlDefault: https://academic.oup.com/journals/ providerName: Oxford University Press
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3NbtNAEF6FwIELggJqKKA9IDhYpvZu_HcsLRUFtXAIUm7Wer1uTVM7st2Q3HgHXoun4EmY_XMaClI5cLHsze5KyXyZnRl_M4PQCypYwFmRuGIcyKLalLpJAY6rENzzCprFHslUs4no5CSeTpNPg8EPmwuzmEVVFS-Xyfy_ihrGQNgydfYfxN1vCgNwD0KHK4gdrjcS_IEQ82OZU-lYIgN1jo5eO29MA2ynqbPLtnNymGebRpw6bHZaN2V3puszODIpUw7D-cGWK6fMDalIw4Wps9XJ6wtWmmifbQpVV93ZbOVUtSTVtZaBoIMUe62Mu9e6QIGhi1qBK6M2LOdN-YVV9aJUz9FGsPUDK861UutZygf1V8NW66O174VsBtGuo7ompEH8nrPVq2kagJo2WlhozazoeIlu0WJVd-CF1zCqFXFomsPoMz3UHcGuHRe6lNZF1cjkuUMwxzO_X3i1MvdvJ2bPY9Rv8Gmqdkjt-lvoNomCRBIMJx-n6yxdqpr_9V-uLyFKd9X6Xbt-w0Qa1qDqr5g8k_vonvFV8J7G2AM0ENUW2rZSXOGXWN3r4Fi7hUbH4IHVjXpRAx_uz0pwh9TTQ3TewxL__PYdAyCxBSTWgMQSkNgCEq8BiQGQ2AASa0DiTUBiDUhsAPkIfT58O9l_55pGHy6nlHQu9fIoJzlNBAg09PJxzMCN9angRMQJ8QuSFR7jocfyhGdg84452KUB8_OIkTBg9DEaVnUlthGOi6xgYFUTcITHfBxl4ADxoAArPC4SUogRcu1vm3JTBV82Y5mlf5blCL3q5891_Ze_ztyRokrBcpXll7nkqfEu9WW174Q8ufE-O-ju-j_xFA275lI8Q3f4oivb5rmC1C_pkLl-
linkProvider	Oxford University Press
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=DeepMerge+%E2%80%93+II.+Building+robust+deep+learning+algorithms+for+merging+galaxy+identification+across+domains&rft.jtitle=Monthly+notices+of+the+Royal+Astronomical+Society&rft.au=%C4%86iprijanovi%C4%87%2C+A&rft.au=Kafkes%2C+D&rft.au=Downey%2C+K&rft.au=Jenkins%2C+S&rft.date=2021-09-01&rft.issn=0035-8711&rft.eissn=1365-2966&rft.volume=506&rft.issue=1&rft.spage=677&rft.epage=691&rft_id=info:doi/10.1093%2Fmnras%2Fstab1677&rft.externalDBID=n%2Fa&rft.externalDocID=10_1093_mnras_stab1677
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0035-8711&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0035-8711&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0035-8711&client=summon