{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,9]],"date-time":"2026-05-09T17:36:35Z","timestamp":1778348195032,"version":"3.51.4"},"reference-count":31,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2023,1,30]],"date-time":"2023-01-30T00:00:00Z","timestamp":1675036800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,1,30]],"date-time":"2023-01-30T00:00:00Z","timestamp":1675036800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100004955","name":"\u00d6sterreichische Forschungsf\u00f6rderungsgesellschaft","doi-asserted-by":"publisher","award":["ML4CPD"],"award-info":[{"award-number":["ML4CPD"]}],"id":[{"id":"10.13039\/501100004955","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Austrian Science Fund"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Eur. Phys. J. Plus"],"abstract":"<jats:title>Abstract<\/jats:title><jats:p>The CRESST experiment employs cryogenic calorimeters for the sensitive measurement of nuclear recoils induced by dark matter particles. The recorded signals need to undergo a careful cleaning process to avoid wrongly reconstructed recoil energies caused by pile-up and read-out artefacts. We frame this process as a time series classification task and propose to automate it with neural networks. With a data set of over one million labeled records from 68 detectors, recorded between 2013 and 2019 by CRESST, we test the capability of four commonly used neural network architectures to learn the data cleaning task. Our best performing model achieves a balanced accuracy of 0.932 on our test set. We show on an exemplary detector that about half of the wrongly predicted events are in fact wrongly labeled events, and a large share of the remaining ones have a context-dependent ground truth. We furthermore evaluate the recall and selectivity of our classifiers with simulated data. The results confirm that the trained classifiers are well suited for the data cleaning task.<\/jats:p>","DOI":"10.1140\/epjp\/s13360-023-03674-2","type":"journal-article","created":{"date-parts":[[2023,1,30]],"date-time":"2023-01-30T07:03:34Z","timestamp":1675062214000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":12,"title":["Towards an automated data cleaning with deep learning in CRESST"],"prefix":"10.1140","volume":"138","author":[{"given":"G.","family":"Angloher","sequence":"first","affiliation":[]},{"given":"S.","family":"Banik","sequence":"additional","affiliation":[]},{"given":"D.","family":"Bartolot","sequence":"additional","affiliation":[]},{"given":"G.","family":"Benato","sequence":"additional","affiliation":[]},{"given":"A.","family":"Bento","sequence":"additional","affiliation":[]},{"given":"A.","family":"Bertolini","sequence":"additional","affiliation":[]},{"given":"R.","family":"Breier","sequence":"additional","affiliation":[]},{"given":"C.","family":"Bucci","sequence":"additional","affiliation":[]},{"given":"J.","family":"Burkhart","sequence":"additional","affiliation":[]},{"given":"L.","family":"Canonica","sequence":"additional","affiliation":[]},{"given":"A.","family":"D\u2019Addabbo","sequence":"additional","affiliation":[]},{"given":"S.","family":"Di\u00a0Lorenzo","sequence":"additional","affiliation":[]},{"given":"L.","family":"Einfalt","sequence":"additional","affiliation":[]},{"given":"A.","family":"Erb","sequence":"additional","affiliation":[]},{"given":"F. v.","family":"Feilitzsch","sequence":"additional","affiliation":[]},{"given":"N. Ferreiro","family":"Iachellini","sequence":"additional","affiliation":[]},{"given":"S.","family":"Fichtinger","sequence":"additional","affiliation":[]},{"given":"D.","family":"Fuchs","sequence":"additional","affiliation":[]},{"given":"A.","family":"Fuss","sequence":"additional","affiliation":[]},{"given":"A.","family":"Garai","sequence":"additional","affiliation":[]},{"given":"V. M.","family":"Ghete","sequence":"additional","affiliation":[]},{"given":"S.","family":"Gerster","sequence":"additional","affiliation":[]},{"given":"P.","family":"Gorla","sequence":"additional","affiliation":[]},{"given":"P. V.","family":"Guillaumon","sequence":"additional","affiliation":[]},{"given":"S.","family":"Gupta","sequence":"additional","affiliation":[]},{"given":"D.","family":"Hauff","sequence":"additional","affiliation":[]},{"given":"M.","family":"Je\u0161kovsk\u00fd","sequence":"additional","affiliation":[]},{"given":"J.","family":"Jochum","sequence":"additional","affiliation":[]},{"given":"M.","family":"Kaznacheeva","sequence":"additional","affiliation":[]},{"given":"A.","family":"Kinast","sequence":"additional","affiliation":[]},{"given":"H.","family":"Kluck","sequence":"additional","affiliation":[]},{"given":"H.","family":"Kraus","sequence":"additional","affiliation":[]},{"given":"M.","family":"Lackner","sequence":"additional","affiliation":[]},{"given":"A.","family":"Langenk\u00e4mper","sequence":"additional","affiliation":[]},{"given":"M.","family":"Mancuso","sequence":"additional","affiliation":[]},{"given":"L.","family":"Marini","sequence":"additional","affiliation":[]},{"given":"L.","family":"Meyer","sequence":"additional","affiliation":[]},{"given":"V.","family":"Mokina","sequence":"additional","affiliation":[]},{"given":"A.","family":"Nilima","sequence":"additional","affiliation":[]},{"given":"M.","family":"Olmi","sequence":"additional","affiliation":[]},{"given":"T.","family":"Ortmann","sequence":"additional","affiliation":[]},{"given":"C.","family":"Pagliarone","sequence":"additional","affiliation":[]},{"given":"L.","family":"Pattavina","sequence":"additional","affiliation":[]},{"given":"F.","family":"Petricca","sequence":"additional","affiliation":[]},{"given":"W.","family":"Potzel","sequence":"additional","affiliation":[]},{"given":"P.","family":"Povinec","sequence":"additional","affiliation":[]},{"given":"F.","family":"Pr\u00f6bst","sequence":"additional","affiliation":[]},{"given":"F.","family":"Pucci","sequence":"additional","affiliation":[]},{"given":"F.","family":"Reindl","sequence":"additional","affiliation":[]},{"given":"D.","family":"Rizvanovic","sequence":"additional","affiliation":[]},{"given":"J.","family":"Rothe","sequence":"additional","affiliation":[]},{"given":"K.","family":"Sch\u00e4ffner","sequence":"additional","affiliation":[]},{"given":"J.","family":"Schieck","sequence":"additional","affiliation":[]},{"given":"D.","family":"Schmiedmayer","sequence":"additional","affiliation":[]},{"given":"S.","family":"Sch\u00f6nert","sequence":"additional","affiliation":[]},{"given":"C.","family":"Schwertner","sequence":"additional","affiliation":[]},{"given":"M.","family":"Stahlberg","sequence":"additional","affiliation":[]},{"given":"L.","family":"Stodolsky","sequence":"additional","affiliation":[]},{"given":"C.","family":"Strandhagen","sequence":"additional","affiliation":[]},{"given":"R.","family":"Strauss","sequence":"additional","affiliation":[]},{"given":"I.","family":"Usherov","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5687-6392","authenticated-orcid":false,"given":"F.","family":"Wagner","sequence":"additional","affiliation":[]},{"given":"M.","family":"Willers","sequence":"additional","affiliation":[]},{"given":"V.","family":"Zema","sequence":"additional","affiliation":[]},{"given":"W.","family":"Waltenberger","sequence":"additional","affiliation":[]},{"name":"CRESST Collaboration","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2023,1,30]]},"reference":[{"key":"3674_CR1","doi-asserted-by":"publisher","unstructured":"N. Planck Collaboration, Y. Aghanim, Akrami et al., Planck 2018 results - vi. cosmological parameters. (2020) https:\/\/doi.org\/10.1051\/0004-6361\/201833910","DOI":"10.1051\/0004-6361\/201833910"},{"key":"3674_CR2","unstructured":"G. Angloher, S. Banik, G. Benato et al., Latest observations on the low energy excess in CRESST-III,\u201d (2022). arXiv:2207.09375"},{"key":"3674_CR3","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.100.102002","volume":"100","author":"AH Abdelhameed","year":"2019","unstructured":"A.H. Abdelhameed, G. Angloher, P. Bauer et al., First results from the CRESST-III low-mass dark matter program. Phys. Rev. D 100, 102002 (2019). https:\/\/doi.org\/10.1103\/PhysRevD.100.102002","journal-title":"Phys. Rev. D"},{"key":"3674_CR4","doi-asserted-by":"crossref","unstructured":"G. Angloher, S. Banik, G. Benato et al., Testing spin-dependent dark matter interactions with lithium aluminate targets in CRESST-III, (2022). arXiv:2207.07640","DOI":"10.1103\/PhysRevD.106.092008"},{"key":"3674_CR5","doi-asserted-by":"publisher","first-page":"917","DOI":"10.1007\/s10618-019-00619-1","volume":"33","author":"H Ismail Fawaz","year":"2019","unstructured":"H. Ismail Fawaz, G. Forestier, J. Weber et al., Deep learning for time series classification: a review. Data Mining and Knowle. Dis. 33, 917\u2013963 (2019). https:\/\/doi.org\/10.1007\/s10618-019-00619-1","journal-title":"Data Mining and Knowle. Dis."},{"issue":"6","key":"3674_CR6","doi-asserted-by":"publisher","first-page":"450","DOI":"10.1140\/epjc\/s10052-019-6869-2","volume":"79","author":"P Holl","year":"2019","unstructured":"P. Holl, L. Hauertmann, B. Majorovits et al., Deep learning based pulse shape discrimination for germanium detectors. Eur. Phys. J. C 79(6), 450 (2019). https:\/\/doi.org\/10.1140\/epjc\/s10052-019-6869-2","journal-title":"Eur. Phys. J. C"},{"issue":"9","key":"3674_CR7","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6471\/ab8e94","volume":"47","author":"CK Khosa","year":"2020","unstructured":"C.K. Khosa, L. Mars, J. Richards, V. Sanz, Convolutional neural networks for direct detection of dark matter. J. Phys. G: Nucl. Part. Phys. 47(9), 095201 (2020). https:\/\/doi.org\/10.1088\/1361-6471\/ab8e94","journal-title":"J. Phys. G: Nucl. Part. Phys."},{"key":"3674_CR8","doi-asserted-by":"publisher","unstructured":"A. Abdulaziz, J. Zhou, A. Di Fulvio et al., Semi-supervised gaussian mixture variational autoencoder for pulse shape discrimination. In: ICASSP 2022 \u2013 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp.\u00a03538\u20133542 (2022). https:\/\/doi.org\/10.1109\/ICASSP43922.2022.9747313","DOI":"10.1109\/ICASSP43922.2022.9747313"},{"key":"3674_CR9","unstructured":"A.J. Z\u00f6ller, Artificial neural network based pulse-shape analysis for cryogenic detectors operated in CRESST-II. Dissertation, Technische Universit\u00e4t M\u00fcnchen, M\u00fcnchen, (2016). http:\/\/mediatum.ub.tum.de\/?id=1303343"},{"issue":"08","key":"3674_CR10","doi-asserted-by":"publisher","first-page":"P08023","DOI":"10.1088\/1748-0221\/13\/08\/p08023","volume":"13","author":"S Delaquis","year":"2018","unstructured":"S. Delaquis, M. Jewell, I. Ostrovskiy et al., Deep neural networks for energy and position reconstruction in EXO-200. J. Instrum. 13(08), P08023\u2013P08023 (2018). https:\/\/doi.org\/10.1088\/1748-0221\/13\/08\/p08023","journal-title":"J. Instrum."},{"key":"3674_CR11","unstructured":"C.M\u00fchlmann, Pulse-shape dicrimination with deep learning in CRESST, (2019). http:\/\/hdl.handle.net\/20.500.12708\/14865"},{"key":"3674_CR12","doi-asserted-by":"publisher","unstructured":"F. Wagner, Machine learning methods for the raw data analysis of cryogenic dark matter experiments, (2020). https:\/\/doi.org\/10.34726\/hss.2020.77322","DOI":"10.34726\/hss.2020.77322"},{"key":"3674_CR13","doi-asserted-by":"publisher","DOI":"10.1007\/s10909-022-02741-9","author":"G Fantini","year":"2022","unstructured":"G. Fantini, A. Armatol, E. Armengaud et al., Machine learning techniques for pile-up rejection in cryogenic calorimeters. J. Low Temp. Phys. (2022). https:\/\/doi.org\/10.1007\/s10909-022-02741-9","journal-title":"J. Low Temp. Phys."},{"key":"3674_CR14","unstructured":"F. Wagner, Nonlinear pile-up separation with lstm neural networks for cryogenic particle detectors, (2021). arXiv:2112.06792"},{"key":"3674_CR15","doi-asserted-by":"publisher","DOI":"10.1007\/s10909-022-02719-7","author":"Y Ichinohe","year":"2022","unstructured":"Y. Ichinohe, S. Yamada, R. Hayakawa et al., Application of deep learning to the evaluation of goodness in the waveform processing of transition-edge sensor calorimeters. J. Low Temp. Phys. (2022). https:\/\/doi.org\/10.1007\/s10909-022-02719-7","journal-title":"J. Low Temp. Phys."},{"issue":"03","key":"3674_CR16","doi-asserted-by":"publisher","first-page":"P03032","DOI":"10.1088\/1748-0221\/16\/03\/p03032","volume":"16","author":"R Huang","year":"2021","unstructured":"R. Huang, E. Armengaud, C. Augier et al., Pulse shape discrimination in CUPID-mo using principal component analysis. J. Instrum. 16(03), P03032 (2021). https:\/\/doi.org\/10.1088\/1748-0221\/16\/03\/p03032","journal-title":"J. Instrum."},{"key":"3674_CR17","doi-asserted-by":"crossref","unstructured":"F. Wagner, D. Bartolot, D. Rizvanovic, et al., Cait: analysis toolkit for cryogenic particle detectors in python, (2022). arXiv:2207.02187","DOI":"10.1007\/s41781-022-00092-4"},{"key":"3674_CR18","doi-asserted-by":"publisher","unstructured":"W. Seidel, G. Forster, W. Christen, et al., Phase transition thermometers with high temperature resolution for calorimetric particle detectors employing dielectric absorbers, https:\/\/doi.org\/10.1016\/0370-2693(90)90388-MPhysics Letters B236 no.\u00a04, 483\u2013487 (1990). https:\/\/www.sciencedirect.com\/science\/article\/pii\/037026939090388M","DOI":"10.1016\/0370-2693(90)90388-M"},{"issue":"1","key":"3674_CR19","doi-asserted-by":"publisher","first-page":"69","DOI":"10.1007\/BF00753837","volume":"100","author":"F Pr\u00f6bst","year":"1995","unstructured":"F. Pr\u00f6bst, M. Frank, S. Cooper et al., Model for cryogenic particle detectors with superconducting phase transition thermometers. J. Low Temp. Phys. 100(1), 69\u2013104 (1995). https:\/\/doi.org\/10.1007\/BF00753837","journal-title":"J. Low Temp. Phys."},{"key":"3674_CR20","doi-asserted-by":"publisher","unstructured":"K. Hornik, M. Stinchcombe, H. White, Multilayer feedforward networks are universal approximators, https:\/\/doi.org\/10.1016\/0893-6080(89)90020-8Neural Networks2 no.\u00a05, 359\u2013366 (1989). https:\/\/www.sciencedirect.com\/science\/article\/pii\/0893608089900208","DOI":"10.1016\/0893-6080(89)90020-8"},{"key":"3674_CR21","doi-asserted-by":"publisher","unstructured":", A high-bias, low-variance introduction to machine learning for physicists, https:\/\/doi.org\/10.1016\/j.physrep.2019.03.001Physics Reports810 1\u2013124 (2019). https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0370157319300766. A high-bias, low-variance introduction to Machine Learning for physicists","DOI":"10.1016\/j.physrep.2019.03.001"},{"key":"3674_CR22","doi-asserted-by":"publisher","unstructured":"S. Hochreiter and J.\u00a0Schmidhuber, Long short-term memory, Neural Comput.9 no.\u00a08, 1735\u20131780 (Nov., 1997). https:\/\/doi.org\/10.1162\/neco.1997.9.8.1735._eprint: https:\/\/direct.mit.edu\/neco\/article-pdf\/9\/8\/1735\/813796\/neco.1997.9.8.1735.pdf","DOI":"10.1162\/neco.1997.9.8.1735."},{"key":"3674_CR23","unstructured":"A. Vaswani, N. Shazeer, N. Parmar, et al., Attention is all you need, In: advances in neural information processing systems, I.\u00a0Guyon, U.\u00a0V. Luxburg, S.\u00a0Bengio, et\u00a0al., eds., vol.\u00a030. Curran Associates, Inc., (2017). https:\/\/proceedings.neurips.cc\/paper\/2017\/file\/3f5ee243547dee91fbd053c1c4a845aa-Paper.pdf"},{"key":"3674_CR24","doi-asserted-by":"crossref","unstructured":"G. Zerveas, S. Jayaraman, D. Patel, et al., A transformer-based framework for multivariate time series representation learning, (2020). arXiv:2010.02803","DOI":"10.1145\/3447548.3467401"},{"key":"3674_CR25","unstructured":"J. Devlin, M.-W. Chang, K. Lee, K. Toutanova, Bert: Pre-training of deep bidirectional transformers for language understanding, (2018). arXiv:1810.04805"},{"key":"3674_CR26","unstructured":"G. Zerveas, Multivariate time series transformer framework. https:\/\/github.com\/gzerveas\/mvts_transformer, (2021)"},{"key":"3674_CR27","unstructured":"A. Paszke, S. Gross, F. Massa, et al., Pytorch: An imperative style, high-performance deep learning library, In: Advances in Neural Information Processing Systems 32, pp.\u00a08024\u20138035. Curran Associates, Inc., (2019). http:\/\/papers.neurips.cc\/paper\/9015-pytorch-an-imperative-style-high-performance-deep-learning-library.pdf"},{"key":"3674_CR28","unstructured":"D. P. Kingma, J. Ba, Adam: A method for stochastic optimization, (2014). arXiv:1412.6980"},{"key":"3674_CR29","unstructured":"L. N. Smith, Cyclical learning rates for training neural networks, (2015). arXiv:1506.01186"},{"key":"3674_CR30","doi-asserted-by":"publisher","unstructured":"M. Carrettoni, O. Cremonesi, Generation of noise time series with arbitrary power spectrum. Communications 181(12), 1982\u20131985 (2010). https:\/\/doi.org\/10.1016\/j.cpc.2010.09.003ComputerPhysicshttps:\/\/www.sciencedirect.com\/science\/article\/pii\/S0010465510003486","DOI":"10.1016\/j.cpc.2010.09.003ComputerPhysics"},{"key":"3674_CR31","unstructured":"A. Baevski, H. Zhou, A. Mohamed, M. Auli, Wav2vec 2.0: A framework for self-supervised learning of speech representations, In: Proceedings of the 34th International Conference on Neural Information Processing Systems, NIPS\u201920. Curran Associates Inc., Red Hook, NY, USA, (2020)"}],"container-title":["The European Physical Journal Plus"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1140\/epjp\/s13360-023-03674-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1140\/epjp\/s13360-023-03674-2\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1140\/epjp\/s13360-023-03674-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,1,30]],"date-time":"2023-01-30T07:04:26Z","timestamp":1675062266000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1140\/epjp\/s13360-023-03674-2"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,1,30]]},"references-count":31,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2023,1]]}},"alternative-id":["3674"],"URL":"https:\/\/doi.org\/10.1140\/epjp\/s13360-023-03674-2","relation":{},"ISSN":["2190-5444"],"issn-type":[{"value":"2190-5444","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,1,30]]},"assertion":[{"value":"1 November 2022","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"2 January 2023","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"30 January 2023","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"On behalf of all authors, the corresponding author states that there is no conflict of interest.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}}],"article-number":"100"}}