CERN Accelerating science

If you experience any problem watching the video, click the download button below
Download Embed
Preprint
Report number arXiv:2408.16612
Title Data Quality Monitoring through Transfer Learning on Anomaly Detection for the Hadron Calorimeters
Author(s)

Asres, Mulugeta Weldezgina (U. Agder, Kristiansand) ; Omlin, Christian Walter (U. Agder, Kristiansand) ; Wang, Long (Maryland U.) ; Parygin, Pavel (Rochester U.) ; Yu, David (Brown U.) ; Dittmann, Jay (Baylor U.) ; Gevorgyan, A. (Yerevan Phys. Inst.) ; Petrosyan, A. (Yerevan Phys. Inst.) ; Tumasyan, A. (Yerevan Phys. Inst.) ; Alves, G.A. (Rio de Janeiro, CBPF) ; Hensel, C. (Rio de Janeiro, CBPF) ; Aldá Júnior, W.L. (Rio de Janeiro State U.) ; Carvalho, W. (Rio de Janeiro State U.) ; Chinellato, J. (Rio de Janeiro State U. ; Saclay) ; De Oliveira Martins, C. (Rio de Janeiro State U.) ; Figueiredo, D. Matos (Rio de Janeiro State U.) ; Mora Herrera, C. (Rio de Janeiro State U.) ; Nogima, H. (Rio de Janeiro State U.) ; Da Silva, W.L. Prado (Rio de Janeiro State U.) ; Tonelli Manganote, E.J. (Rio de Janeiro State U.) ; Vilela Pereira, A. (Rio de Janeiro State U.) ; Finger, M. (Charles U.) ; Finger, Jr.M. (Charles U.) ; Adamov, G. (GTU, Tbilisi) ; Tsamalaidze, Z. (GTU, Tbilisi ; Tbilisi State U.) ; Borras, K. (DESY ; CERN) ; Campbell, A. (DESY) ; Engelke, F. (DESY ; CERN) ; Krücker, D. (DESY) ; Martens, I. (DESY) ; Wiens, L. (DESY ; CERN) ; Csanád, M. (Eotvos U.) ; Feherkuti, A. (Eotvos U.) ; Lökös, S. (Eotvos U.) ; Pásztor, G. (Eotvos U.) ; Surányi, O. (Eotvos U.) ; Veres, G.I. (Eotvos U.) ; Kansal, B. (IISER, Pune) ; Sharma, S. (IISER, Pune) ; Beri, S.B. (Panjab U., Chandigarh) ; Bhawandeep, B. (Panjab U., Chandigarh) ; Chawla, R. (Panjab U., Chandigarh) ; Kalsi, A. (Panjab U., Chandigarh) ; Kaur, A. (Panjab U., Chandigarh) ; Kaur, M. (Panjab U., Chandigarh) ; Walia, G. (Panjab U., Chandigarh) ; Bhattacharya, S. (Saha Inst.) ; Ghosh, S. (Saha Inst.) ; Nandan, S. (Saha Inst.) ; Purohit, A. (Saha Inst.) ; Sharan, M. (Saha Inst.) ; Banerjee, S. (Tata Inst.) ; Bhattacharya, S. (Tata Inst.) ; Chatterjee, S. (Tata Inst.) ; Das, P. (Tata Inst.) ; Guchait, M. (Tata Inst.) ; Jain, S. (Tata Inst.) ; Kumar, S. (Tata Inst.) ; Maity, M. (Tata Inst.) ; Majumder, G. (Tata Inst.) ; Mazumdar, K. (Tata Inst.) ; Patil, M. (Tata Inst.) ; Sarkar, T. (Tata Inst.) ; Sekmen, S. (Kyungpook Natl. U. ; CERN) ; Juodagalvis, A. (Vilnius U.) ; Agyel, D. (Mersin U.) ; Boran, F. (Mersin U.) ; Damarseckin, S. (Mersin U.) ; Demiroglu, Z.S. (Mersin U.) ; Dölek, F. (Mersin U.) ; Dumanoglu, I. (Mersin U. ; Near East U.) ; Eskut, E. (Mersin U.) ; Gokbulut, G. (Mersin U.) ; Guler, Y. (Mersin U. ; Konya Technical U.) ; Gurpinar Guler, E. (Mersin U. ; Konya Technical U.) ; Işik, C. (Mersin U.) ; Kangal, E.E. (Mersin U.) ; Kara, O. (Mersin U.) ; Kayis Topaksu, A. (Mersin U.) ; Kiminsu, U. (Mersin U.) ; Onengut, G. (Mersin U.) ; Ozdemir, K. (Mersin U. ; Izmir U. Economics, Izmir) ; Pinar, E. (Mersin U.) ; Polatoz, A. (Mersin U.) ; Simsek, A.E. (Mersin U.) ; Tali, B. (Mersin U. ; Adiyaman U.) ; Tok, U.G. (Mersin U.) ; Turkcapar, S. (Mersin U.) ; Uslan, E. (Mersin U.) ; Zorbakir, I.S. (Mersin U.) ; Bilin, B. (Middle East Tech. U., Ankara ; CERN) ; Karapinar, G. (Middle East Tech. U., Ankara ; Istanbul U.) ; Guler, A. Murat (Middle East Tech. U., Ankara) ; Ocalan, K. (Middle East Tech. U., Ankara ; Necmettin Erbakan U., Konya) ; Yalvac, M. (Middle East Tech. U., Ankara ; Yozgat Bozok U.) ; Zeyrek, M. (Middle East Tech. U., Ankara) ; Akgun, B. (Bogazici U.) ; Atakisi, I.O. (Bogazici U. ; Marmara U.) ; Gülmez, E. (Bogazici U.) ; Kaya, M. (Bogazici U. ; Marmara U.) ; Kaya, O. (Bogazici U. ; Yildiz Tech. U.) ; Tekten, S. (Bogazici U. ; Uppsala U.) ; Yetkin, E.A. (Bogazici U. ; Istanbul U.) ; Yetkin, T. (Bogazici U. ; Yildiz Tech. U.) ; Cakir, A. (Istanbul, Tech. U.) ; Cankocak, K. (Istanbul, Tech. U. ; Near East U.) ; Sen, S. (Istanbul, Tech. U. ; Hacettepe U.) ; Aydilek, O. (Istanbul U.) ; Cerci, S. (Istanbul U. ; Adiyaman U.) ; Hacisahinoglu, B. (Istanbul U.) ; Hos, I. (Istanbul U. ; Istanbul, Tech. U.) ; Isildak, B. (Istanbul U. ; Yildiz Tech. U.) ; Kaynak, B. (Istanbul U.) ; Ozkorucuklu, S. (Istanbul U.) ; Potok, O. (Istanbul U.) ; Sert, H. (Istanbul U.) ; Simsek, C. (Istanbul U.) ; Sunar Cerci, D. (Istanbul U. ; Adiyaman U.) ; Zorbilmez, C. (Istanbul U.) ; Boyarintsev, A. (Kharkov, Single Crystals Res. Inst.) ; Grynyov, B. (Kharkov, Single Crystals Res. Inst.) ; Levchuk, L. (Kharkov Natl. U.) ; Popov, V. (Kharkov Natl. U.) ; Sorokin, P. (Kharkov Natl. U.) ; Flacher, H. (U. Bristol (main)) ; Abdullin, S. (Baylor U., Math. Dept.) ; Caraway, B. (Baylor U., Math. Dept.) ; Hatakeyama, K. (Baylor U., Math. Dept.) ; Kanuganti, A.R. (Baylor U., Math. Dept.) ; McMaster, B. (Baylor U., Math. Dept.) ; Saunders, M. (Baylor U., Math. Dept.) ; Wilson, J. (Baylor U., Math. Dept.) ; Buccilli, A. (Alabama U. ; San Francisco State U.) ; Bunin, P. (Alabama U. ; Higher Sch. of Economics, Moscow) ; Cooper, S.I. (Alabama U.) ; Henderson, C. (Alabama U. ; Cincinnati U.) ; Perez, C.U. (Alabama U.) ; Rumerio, P. (Alabama U. ; Turin U.) ; Cosby, C. (Boston U.) ; Demiragli, Z. (Boston U.) ; Gastler, D. (Boston U.) ; Hazen, E. (Boston U.) ; Rohlf, J. (Boston U.) ; Hadley, M. (Brown U. ; Fermilab) ; Heintz, U. (Brown U. ; Fermilab) ; Kwon, T. (Brown U. ; Fermilab) ; Laird, E. (Brown U. ; Fermilab) ; Landsberg, G. (Brown U. ; Fermilab) ; Lau, K.T. (Brown U. ; Fermilab) ; Yan, X. (Brown U. ; Fermilab) ; Mao, Z. (Brown U. ; Fermilab) ; Gary, J.W. (UC, Riverside (main)) ; Karapostoli, G. (UC, Riverside (main) ; Natl. Tech. U., Athens) ; Long, O.R. (UC, Riverside (main)) ; Bhandari, R. (UC, Santa Barbara) ; Heller, R. (UC, Santa Barbara) ; Stuart, D. (UC, Santa Barbara) ; Yoo, J. (UC, Santa Barbara ; Korea U.) ; Chen, Y. (Caltech, Kellogg Lab ; MIT) ; Duarte, J. (Caltech, Kellogg Lab) ; Lawhorn, J.M. (Caltech, Kellogg Lab) ; Spiropulu, M. (Caltech, Kellogg Lab) ; Apresyan, A. (Fermilab) ; Apyan, A. (Fermilab ; Brandeis U.) ; Banerjee, S. (Fermilab ; Wisconsin U., Madison) ; Chlebana, F. (Fermilab) ; Feng, Y. (Fermilab) ; Freeman, J. (Fermilab) ; Green, D. (Fermilab) ; Kwok, K.H.M. (Fermilab) ; Hirschauer, J. (Fermilab) ; Joshi, U. (Fermilab) ; Lincoln, D. (Fermilab) ; Los, S. (Fermilab) ; Madrid, C. (Fermilab) ; Pastika, N. (Fermilab) ; Pedro, K. (Fermilab) ; Spalding, W.J. (Fermilab) ; Tkaczyk, S. (Fermilab) ; Linn, S. (Florida Intl. U.) ; Markowitz, P. (Florida Intl. U.) ; Hagopian, V. (Yerevan Phys. Inst. ; Florida State U.) ; Kolberg, T. (Yerevan Phys. Inst. ; Florida State U.) ; Martinez, G. (Yerevan Phys. Inst. ; Florida State U.) ; Viazlo, O. (Yerevan Phys. Inst. ; Florida State U.) ; Hohlmann, M. (Florida Inst. Tech.) ; Verma, R. Kumar (Florida Inst. Tech.) ; Noonan, D. (Florida Inst. Tech.) ; Yumiceva, F. (Florida Inst. Tech. ; Unlisted, US) ; Alhusseini, M. (Iowa U.) ; Bilki, B. (Iowa U.) ; Blend, D. (Iowa U.) ; Dilsiz, K. (Iowa U. ; Bingöl U.) ; Emediato, L. (Iowa U.) ; Gandrajula, R.P. (Iowa U.) ; Herrmann, M. (Iowa U.) ; Köseyan, O.K. (Iowa U.) ; Merlo, J.-P. (Iowa U.) ; Mestvirishvili, A. (Iowa U. ; GTU, Tbilisi) ; Miller, M. (Iowa U.) ; Ogul, H. (Iowa U. ; Sakarya U.) ; Onel, Y. (Iowa U.) ; Penzo, A. (Iowa U.) ; Southwick, D. (Iowa U.) ; Tiras, E. (Iowa U. ; Erciyes U.) ; Wetzel, J. (Iowa U.) ; Al-bataineh, A. (Kansas U. ; Yarmouk U.) ; Bowen, J. (Kansas U. ; Saclay) ; Le Mahieu, C. (Kansas U.) ; Marquez, J. (Kansas U.) ; McBrayer, W. (Kansas U.) ; Murray, M. (Kansas U.) ; Nickel, M. (Kansas U.) ; Popescu, S. (Kansas U. ; Bucharest, IFIN-HH) ; Smith, C. (Kansas U.) ; Wang, Q. (Kansas U.) ; Kaadze, K. (Kansas State U.) ; Kim, D. (Kansas State U.) ; Maravin, Y. (Kansas State U.) ; Mohammadi, A. (Kansas State U. ; Wisconsin U., Madison) ; Natoli, J. (Kansas State U.) ; Roy, D. (Kansas State U.) ; Saini, L.K. (Kansas State U. ; Saclay) ; Adams, E. (Maryland U., College Park) ; Baden, A. (Maryland U., College Park) ; Baron, O. (Maryland U., College Park) ; Belloni, A. (Maryland U., College Park) ; Bethani, A. (Maryland U., College Park) ; Chen, Y-M (Maryland U., College Park) ; Eno, S.C. (Maryland U., College Park) ; Ferraioli, C. (Maryland U., College Park ; Saclay) ; Grassi, T. (Maryland U., College Park) ; Hadley, N.J. (Maryland U., College Park) ; Kellogg, R.G. (Maryland U., College Park) ; Koeth, T. (Maryland U., College Park) ; Lai, Y. (Maryland U., College Park) ; Lascio, S. (Maryland U., College Park) ; Mignerey, A.C. (Maryland U., College Park) ; Nabili, S. (Maryland U., College Park) ; Palmer, C. (Maryland U., College Park) ; Papageorgakis, C. (Maryland U., College Park) ; Seidel, M. (Maryland U., College Park ; Latvia U., ISSP) ; Wong, K. (Maryland U., College Park) ; D'Alfonso, M. (MIT) ; Hu, M. (MIT) ; Crossman, B. (Minnesota U.) ; Hiltbrand, J. (Minnesota U.) ; Krohn, M. (Minnesota U.) ; Mans, J. (Minnesota U.) ; Revering, M. (Minnesota U.) ; Strobbe, N. (Minnesota U.) ; Heering, A. (Notre Dame U.) ; Musienko, Y. (Notre Dame U. ; Higher Sch. of Economics, Moscow) ; Ruchti, R. (Notre Dame U.) ; Wayne, M. (Notre Dame U.) ; Chung, W. (Princeton U.) ; Kopp, G. (Princeton U.) ; Mei, K. (Princeton U.) ; Tully, C. (Princeton U.) ; Bodek, A. (Rochester U.) ; de Barbaro, P. (Rochester U.) ; Fallon, C. (Rochester U.) ; Galanti, M. (Rochester U.) ; Garcia-Bellido, A. (Rochester U.) ; Khukhunaishvili, A. (Rochester U.) ; Tan, C-L (Rochester U.) ; Taus, R. (Rochester U.) ; Vishnevskiy, D. (Rochester U.) ; Zielinski, M. (Rochester U.) ; Chiarito, B. (Rutgers U., Piscataway) ; Chou, J.P. (Rutgers U., Piscataway) ; Thayil, S.A. (Rutgers U., Piscataway) ; Wang, H. (Rutgers U., Piscataway) ; Akchurin, N. (Texas Tech.) ; Damgov, J. (Texas Tech.) ; De Guio, F. (Texas Tech. ; INFN, Milan Bicocca) ; Kunori, S. (Texas Tech.) ; Lamichhane, K. (Texas Tech.) ; Lee, S.W. (Texas Tech.) ; Mengke, T. (Texas Tech.) ; Muthumuni, S. (Texas Tech.) ; Undleeb, S. (Texas Tech.) ; Volobouev, I. (Texas Tech.) ; Wang, Z. (Texas Tech.) ; Whitbeck, A. (Texas Tech.) ; Cummings, G. (Virginia U.) ; Goadhouse, S. (Virginia U.) ; Hakala, J. (Virginia U.) ; Hirosky, R. (Virginia U.) ; Winn, D. (Fairfield U.) ; Alexakhin, V. (Higher Sch. of Economics, Moscow) ; Andreev, V. (Higher Sch. of Economics, Moscow) ; Andreev, Y. (Higher Sch. of Economics, Moscow) ; Azarkin, M. (Higher Sch. of Economics, Moscow) ; Belyaev, A. (Higher Sch. of Economics, Moscow) ; Bitioukov, S. (Higher Sch. of Economics, Moscow) ; Boos, E. (Higher Sch. of Economics, Moscow) ; Bychkova, O. (Higher Sch. of Economics, Moscow) ; Chadeeva, M. (Higher Sch. of Economics, Moscow) ; Chekhovsky, V. (Higher Sch. of Economics, Moscow) ; Chistov, R. (Higher Sch. of Economics, Moscow) ; Danilov, M. (Higher Sch. of Economics, Moscow) ; Demianov, A. (Higher Sch. of Economics, Moscow) ; Dermenev, A. (Higher Sch. of Economics, Moscow) ; Dubinin, M. (Higher Sch. of Economics, Moscow ; Caltech) ; Dudko, L. (Higher Sch. of Economics, Moscow) ; Elumakhov, D. (Higher Sch. of Economics, Moscow) ; Epshteyn, V. (Higher Sch. of Economics, Moscow) ; Ershov, Y. (Higher Sch. of Economics, Moscow) ; Ershov, A. (Higher Sch. of Economics, Moscow) ; Gavrilov, V. (Higher Sch. of Economics, Moscow) ; Gribushin, A. (Higher Sch. of Economics, Moscow) ; Kalinin, A. (Higher Sch. of Economics, Moscow ; Maryland U., College Park) ; Kaminskiy, A. (Higher Sch. of Economics, Moscow) ; Karneyeu, A. (Higher Sch. of Economics, Moscow) ; Khein, L. (Higher Sch. of Economics, Moscow) ; Kirakosyan, M. (Higher Sch. of Economics, Moscow) ; Klyukhin, V. (Higher Sch. of Economics, Moscow) ; Kodolova, O. (Higher Sch. of Economics, Moscow ; Yerevan State U.) ; Krychkine, V. (Higher Sch. of Economics, Moscow) ; Kurenkov, A. (Higher Sch. of Economics, Moscow) ; Litomin, A. (Higher Sch. of Economics, Moscow) ; Lychkovskaya, N. (Higher Sch. of Economics, Moscow) ; Makarenko, V. (Higher Sch. of Economics, Moscow) ; Mandrik, P. (Higher Sch. of Economics, Moscow) ; Obraztsov, S. (Higher Sch. of Economics, Moscow) ; Oskin, A. (Higher Sch. of Economics, Moscow) ; Parygin, P. (Higher Sch. of Economics, Moscow ; Rochester U.) ; Petrov, V. (Higher Sch. of Economics, Moscow) ; Petrushanko, S. (Higher Sch. of Economics, Moscow) ; Polikarpov, S. (Higher Sch. of Economics, Moscow) ; Popova, E. (Higher Sch. of Economics, Moscow ; Rochester U.) ; Rusinov, V. (Higher Sch. of Economics, Moscow) ; Ryutin, R. (Higher Sch. of Economics, Moscow) ; Savrin, V. (Higher Sch. of Economics, Moscow) ; Selivanova, D. (Higher Sch. of Economics, Moscow) ; Smirnov, V. (Higher Sch. of Economics, Moscow) ; Snigirev, A. (Higher Sch. of Economics, Moscow) ; Sobol, A. (Higher Sch. of Economics, Moscow) ; Stepennov, A. (Higher Sch. of Economics, Moscow ; Cyprus U. ; Cyprus Inst.) ; Tarkovskii, E. (Higher Sch. of Economics, Moscow) ; Terkulov, A. (Higher Sch. of Economics, Moscow) ; Tlisova, I. (Higher Sch. of Economics, Moscow) ; Tolochek, R. (Higher Sch. of Economics, Moscow) ; Toms, M. (Higher Sch. of Economics, Moscow ; KIT, Karlsruhe) ; Toropin, A. (Higher Sch. of Economics, Moscow) ; Troshin, S. (Higher Sch. of Economics, Moscow) ; Volkov, A. (Higher Sch. of Economics, Moscow) ; Yuldashev, B. (Higher Sch. of Economics, Moscow) ; Zarubin, A. (Higher Sch. of Economics, Moscow) ; Zhokin, A. (Higher Sch. of Economics, Moscow)

Imprint 2024-08-29
Number of pages 28
Note 28 pages, 15 figures, and 9 tables
Subject category cs.LG ; Computing and Computers
Accelerator/Facility, Experiment CERN LHC ; CMS
Abstract The proliferation of sensors brings an immense volume of spatio-temporal (ST) data in many domains for various purposes, including monitoring, diagnostics, and prognostics applications. Data curation is a time-consuming process for a large volume of data, making it challenging and expensive to deploy data analytics platforms in new environments. Transfer learning (TL) mechanisms promise to mitigate data sparsity and model complexity by utilizing pre-trained models for a new task. Despite the triumph of TL in fields like computer vision and natural language processing, efforts on complex ST models for anomaly detection (AD) applications are limited. In this study, we present the potential of TL within the context of AD for the Hadron Calorimeter of the Compact Muon Solenoid experiment at CERN. We have transferred the ST AD models trained on data collected from one part of a calorimeter to another. We have investigated different configurations of TL on semi-supervised autoencoders of the ST AD models -- transferring convolutional, graph, and recurrent neural networks of both the encoder and decoder networks. The experiment results demonstrate that TL effectively enhances the model learning accuracy on a target subdetector. The TL achieves promising data reconstruction and AD performance while substantially reducing the trainable parameters of the AD models. It also improves robustness against anomaly contamination in the training data sets of the semi-supervised AD models.
Other source Inspire
Copyright/License preprint: (License: CC BY 4.0)



 


 Journalen skapades 2024-08-31, och modifierades senast 2024-11-11


Fulltext:
2408.16612 - Download fulltextPDF
16056c6ec01589b6976379b43d47c384 - Download fulltextPDF
External link:
Download fulltextFermilab Library Server