{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,26]],"date-time":"2025-11-26T16:33:05Z","timestamp":1764174785086,"version":"3.37.3"},"reference-count":25,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2020,2,23]],"date-time":"2020-02-23T00:00:00Z","timestamp":1582416000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2020,2,23]],"date-time":"2020-02-23T00:00:00Z","timestamp":1582416000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Comput Softw Big Sci"],"published-print":{"date-parts":[[2020,12]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Applications of neural networks to data analyses in natural sciences are complicated by the fact that many inputs are subject to systematic uncertainties. To control the dependence of the neural network function to variations of the input space within these systematic uncertainties, several methods have been proposed. In this work, we propose a new approach of training the neural network by introducing penalties on the variation of the neural network output directly in the loss function. This is achieved at the cost of only a small number of additional hyperparameters. It can also be pursued by treating all systematic variations in the form of statistical weights. The proposed method is demonstrated with a simple example, based on pseudo-experiments, and by a more complex example from high-energy particle physics.<\/jats:p>","DOI":"10.1007\/s41781-020-00037-9","type":"journal-article","created":{"date-parts":[[2020,2,23]],"date-time":"2020-02-23T04:16:02Z","timestamp":1582431362000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":23,"title":["Reducing the Dependence of the Neural Network Function to Systematic Uncertainties in the Input Space"],"prefix":"10.1007","volume":"4","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4485-2972","authenticated-orcid":false,"given":"Stefan","family":"Wunsch","sequence":"first","affiliation":[]},{"given":"Simon","family":"J\u00f6rger","sequence":"additional","affiliation":[]},{"given":"Roger","family":"Wolf","sequence":"additional","affiliation":[]},{"given":"G\u00fcnter","family":"Quast","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2020,2,23]]},"reference":[{"issue":"04","key":"37_CR1","doi-asserted-by":"publisher","first-page":"P04008","DOI":"10.1088\/1748-0221\/11\/04\/P04008","volume":"11","author":"The ATLAS collaboration","year":"2016","unstructured":"The ATLAS collaboration (2016) Performance of $$b$$-jet identification in the ATLAS experiment. JINST 11(04):P04008","journal-title":"JINST"},{"issue":"05","key":"37_CR2","doi-asserted-by":"publisher","first-page":"P05011","DOI":"10.1088\/1748-0221\/13\/05\/P05011","volume":"13","author":"The CMS collaboration","year":"2018","unstructured":"The CMS collaboration (2018) Identification of heavy-flavour jets with the CMS detector in pp collisions at 13 TeV. JINST 13(05):P05011","journal-title":"JINST"},{"issue":"5","key":"37_CR3","doi-asserted-by":"publisher","first-page":"295","DOI":"10.1140\/epjc\/s10052-016-4110-0","volume":"76","author":"The ATLAS collaboration","year":"2016","unstructured":"The ATLAS collaboration (2016) Reconstruction of hadronic decay products of tau leptons with the ATLAS experiment. Eur Phys J C 76(5):295","journal-title":"Eur Phys J C"},{"issue":"10","key":"37_CR4","doi-asserted-by":"publisher","first-page":"P10005","DOI":"10.1088\/1748-0221\/13\/10\/P10005","volume":"13","author":"The CMS collaboration","year":"2018","unstructured":"The CMS collaboration (2018) Performance of reconstruction and identification of $$\\tau$$ leptons decaying to hadrons and $$\\nu _\\tau$$ in pp collisions at $$\\sqrt{s}=$$ 13 TeV. JINST 13(10):P10005","journal-title":"JINST"},{"key":"37_CR5","doi-asserted-by":"publisher","first-page":"173","DOI":"10.1016\/j.physletb.2018.07.035","volume":"784","author":"The ATLAS collaboration","year":"2018","unstructured":"The ATLAS collaboration (2018) Observation of Higgs boson production in association with a top quark pair at the LHC with the ATLAS detector. Phys Lett B 784:173\u2013191","journal-title":"Phys Lett B"},{"key":"37_CR6","doi-asserted-by":"publisher","first-page":"59","DOI":"10.1016\/j.physletb.2018.09.013","volume":"786","author":"The ATLAS collaboration","year":"2018","unstructured":"The ATLAS collaboration (2018) Observation of $$H \\rightarrow b\\bar{b}$$ decays and $$VH$$ production with the ATLAS detector. Phys Lett B 786:59\u201386","journal-title":"Phys Lett B"},{"issue":"23","key":"37_CR7","doi-asserted-by":"publisher","first-page":"231801","DOI":"10.1103\/PhysRevLett.120.231801","volume":"120","author":"The CMS collaboration","year":"2018","unstructured":"The CMS collaboration (2018) Observation of $$\\text{t}\\overline{{\\text{t}}}$$H production. Phys Rev Lett 120(23):231801","journal-title":"Phys Rev Lett"},{"issue":"12","key":"37_CR8","doi-asserted-by":"publisher","first-page":"121801","DOI":"10.1103\/PhysRevLett.121.121801","volume":"121","author":"The CMS collaboration","year":"2018","unstructured":"The CMS collaboration (2018) Observation of Higgs boson decay to bottom quarks. Phys Rev Lett 121(12):121801","journal-title":"Phys Rev Lett"},{"unstructured":"The CMS Collaboration (2019) Measurement of Higgs boson production and decay to the $$\\tau \\tau$$ final state. Technical Report CMS-PAS-HIG-18-032, CERN, Geneva","key":"37_CR9"},{"issue":"7","key":"37_CR10","doi-asserted-by":"publisher","first-page":"074034","DOI":"10.1103\/PhysRevD.96.074034","volume":"96","author":"C Shimmin","year":"2017","unstructured":"Shimmin C et al (2017) Decorrelated jet substructure tagging using adversarial neural networks. Phys Rev D 96(7):074034","journal-title":"Phys Rev D"},{"unstructured":"Goodfellow IJ, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y (2014) Generative adversarial networks, arXiv:1406.2661","key":"37_CR11"},{"unstructured":"Louppe G,  Kagan M, Cranmer K, et\u00a0al (2017) Learning to pivot with adversarial networks. In: Advances in neural information processing systems, pp 981\u2013990","key":"37_CR12"},{"key":"37_CR13","doi-asserted-by":"publisher","first-page":"170","DOI":"10.1016\/j.cpc.2019.06.007","volume":"244","author":"P De Castro","year":"2019","unstructured":"De Castro P, Dorigo T (2019) INFERNO: inference-aware neural optimisation. Comput Phys Commun 244:170\u2013179","journal-title":"Comput Phys Commun"},{"issue":"8","key":"37_CR14","doi-asserted-by":"publisher","first-page":"083004","DOI":"10.1103\/PhysRevD.97.083004","volume":"97","author":"T Charnock","year":"2018","unstructured":"Charnock T, Lavaux G, Wandelt BD (2018) Automatic physical inference with information maximizing neural networks. Phys Rev D 97(8):083004","journal-title":"Phys Rev D"},{"key":"37_CR15","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1016\/j.nima.2019.03.088","volume":"930","author":"LG Xia","year":"2019","unstructured":"Xia LG (2019) QBDT, a new boosting decision tree method with systematical uncertainties into training for High Energy Physics. Nucl Instrum Method A 930:15\u201326","journal-title":"Nucl Instrum Method A"},{"unstructured":"Elwood A, Kr\u00fccker D (2018) Direct optimisation of the discovery significance when training neural networks to search for new physics in particle colliders. Technical report, Deutsches Elektron Synchrotron (DESY)","key":"37_CR16"},{"unstructured":"Estrade V, Germain C, Guyon I, Rousseau (2018) Systematics aware learning: a case study in High Energy Physics. In: ESANN 2018\u201426th European symposium on artificial neural networks, Bruges, Belgium","key":"37_CR17"},{"key":"37_CR18","first-page":"895","volume-title":"Advances in neural information processing systems","author":"P Simard","year":"1992","unstructured":"Simard P, Victorri B, LeCun Y, Denker J (1992) Tangent prop\u2014a formalism for specifying selected invariances in an adaptive network. In: Moody JE, Hanson SJ, Lippmann RP (eds) Advances in neural information processing systems, vol 4. Morgan-Kaufmann, Burlington, pp 895\u2013903"},{"unstructured":"Glorot X et\u00a0al (2011) Deep sparse rectifier neural networks. In: Proceedings of the fourteenth international conference on artificial intelligence and statistics, pp 315\u2013323","key":"37_CR19"},{"unstructured":"Glorot X, Bengio Y (2010) Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the thirteenth international conference on artificial intelligence and statistics, pp 249\u2013256","key":"37_CR20"},{"unstructured":"Kingma D, Ba J (2014) Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980","key":"37_CR21"},{"unstructured":"Adam-Bourdarios C, Cowan G, Germain C, Guyon I, K\u00e9gl B, Rousseau D (2014) The Higgs boson machine learning challenge. In: NIPS 2014 workshop on high-energy physics and machine learning, vol 42 of JMLR: workshop and conference proceedings, Montreal, Canada, p 37","key":"37_CR22"},{"key":"37_CR23","doi-asserted-by":"publisher","first-page":"072001","DOI":"10.1103\/PhysRevD.99.072001","volume":"99","author":"M Aaboud","year":"2019","unstructured":"Aaboud M et al (2019) Cross-section measurements of the Higgs boson decaying into a pair of $$\\tau$$-leptons in proton-proton collisions at $$\\sqrt{s}=13$$ TeV with the ATLAS detector. Phys Rev D 99:072001","journal-title":"Phys Rev D"},{"unstructured":"The ATLAS and CMS Collaborations (2011) Procedure for the LHC Higgs boson search combination in summer 2011. Technical report, ATL-PHYS-PUB-2011-011, CMS NOTE 2011\/005","key":"37_CR24"},{"issue":"1","key":"37_CR25","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1007\/s41781-018-0012-1","volume":"2","author":"S Wunsch","year":"2018","unstructured":"Wunsch S, Friese R, Wolf R, Quast G (2018) Identifying the relevant dependencies of the neural network response on characteristics of the input space. Comput Softw Big Sci 2(1):5","journal-title":"Comput Softw Big Sci"}],"container-title":["Computing and Software for Big Science"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/link.springer.com\/content\/pdf\/10.1007\/s41781-020-00037-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/link.springer.com\/article\/10.1007\/s41781-020-00037-9\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/link.springer.com\/content\/pdf\/10.1007\/s41781-020-00037-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,2,22]],"date-time":"2021-02-22T00:18:48Z","timestamp":1613953128000},"score":1,"resource":{"primary":{"URL":"http:\/\/link.springer.com\/10.1007\/s41781-020-00037-9"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,2,23]]},"references-count":25,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2020,12]]}},"alternative-id":["37"],"URL":"https:\/\/doi.org\/10.1007\/s41781-020-00037-9","relation":{},"ISSN":["2510-2036","2510-2044"],"issn-type":[{"type":"print","value":"2510-2036"},{"type":"electronic","value":"2510-2044"}],"subject":[],"published":{"date-parts":[[2020,2,23]]},"assertion":[{"value":"29 July 2019","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"11 February 2020","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"23 February 2020","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Compliance with Ethical Standards"}},{"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":"5"}}