Machine learning method for enforcing variable independence in background estimation with LHC data: ABCDisCoTEC
Yazarlar (150)
Aram Hayrapetyan
Vladimir Makarenko
Armen Tumasyan
Wolfgang Adam
Janik Walter Andrejkovic
Lisa Benato
Thomas Bergauer
Konstantinos Damanakis
Marko Dragicevic
Cristina Giordano
Priya Sajid Hussain
Manfred Jeitler
Natascha Krammer
Prof. Dr. Hasan OĞUL Sinop Üniversitesi, Türkiye
Dietrich Liko
Ivan Mikulec
Jochen Schieck
Robert Schöfbeck
Dennis Schwarz
Maryam Shooshtari
Mangesh Sonawane
Wolfgang Waltenberger
Claudia-Elisabeth Wulz
Tahys Janssen
Hyejin Kwon
Thomas van Laer
Pierre van Mechelen
Jas Bierkens
Nordin Breugelmans
Jorgen d'Hondt
Soumya Dansana
Alexandre de Moor
Martin Delcourt
Felix Heyen
Yanwen Hong
Pavlo Kashko
Steven Lowette
Inna Makarenko
Denise Müller
Juhee Song
Stefaan Tavernier
Michael Tytgat
Gerrit Patrick van Onsem
Senne van Putte
David Vannerom
Bugra Bilin
Barbara Clerbaux
Aloke Kumar Das
Isabelle de Bruyn
Gilles de Lentdecker
Hugues Evard
Laurent Favart
Paraskevas Gianneios
Ali Khalilzadeh
Fakhri Alam Khan
Andrea Malara
Muhammad Aamir Shahzad
Laurent Thomas
Max Vanden Bemden
Pascal Vanlaer
Fengwangdong Zhang
Maarten de Coen
Didar Dobur
Gul Gokbulut
Joscha Knolle
Luka Lambrecht
David Marckx
Kirill Skovpen
Niels van den Bossche
Jan van der Linden
Jules Vandenbroeck
Liam Wezenbeek
Samuel Bein
Anna Benecke
Agni Bethani
Giacomo Bruno
Alessandra Cappati
Jerome de Favereau de Jeneret
Christophe Delaere
Andrea Giammanco
Ahmet Oguz Guzel
Vincent Lemaitre
Jindrich Lidrych
Paola Mastrapasqua
Semra Turkcapar
Gilvan Alves
Eduardo Coelho
Carsten Hensel
Thales Menezes de Oliveira
Clemencia Mora Herrera
Patricia Rebello Teles
Mariana Soeiro
Antonio Vilela Pereira
Walter Luiz Aldá Júnior
Mapse Barroso Ferreira Filho
Helena Brandao Malbouisson
Wagner Carvalho
Jose Chinellato
Matheus Costa Reis
Eliza Melo da Costa
Gustavo Gil da Silveira
Dilson de Jesus Damiao
Sandro Fonseca de Souza
Raphael Gomes de Souza
Silas Jesus
Tulio Laux Kuhn
Matheus Macedo
Kevin Mota Amarilo
Luiz Mundim
Helio Nogima
Joao Pedro Pinheiro
Alberto Santoro
Andre Sznajder
Mauricio Thiel
Felipe Torres da Silva de Araujo
Cesar Augusto Bernardes
Luigi Calligaris
Thiago Tomei
Eduardo de Moraes Gregores
Bruno Lopes da Costa
Isabela Maietto Silverio
Pedro G Mercadante
Sergio F Novaes
Breno Orzari
Sandra Padula
Valerie Scheurer
Aleksandar Aleksandrov
Georgy Antchev
Petar Danev
Roumyana Hadjiiska
Plamen Iaydjiev
Milena Misheva
Mariana Shopova
Georgi Sultanov
Anton Dimitrov
Leander Litov
Borislav Pavlov
Peicho Petkov
Anton Petrov
Elton Shumka
Sumit Keshri
David Nicolas Laroze Navarrete
Shalini Thakur
William Brooks
Tongguang Cheng
Tahir Javaid
Li Yuan
Zhen Hu
Zhengchen Liang
Jinfeng Liu
Sinop Üniversitesi
Sinop Üniversitesi
| Makale Türü | Özgün Makale (Uluslararası alan indekslerindeki dergilerde yayınlanan tam makale) | ||
| Dergi Adı | |||
| Makale Dili | – | Basım Tarihi | 07-2025 |
| Makale Linki | https://hal.science/hal-05153769/ | ||
| UAK Araştırma Alanları |
Nükleer Fizik
|
||
| Özet |
| A novel solution is presented for the problem of estimating the backgrounds of a signal search using observed data while simultaneously maximizing the sensitivity of the search to the signal. The ``ABCD method'' provides a reliable framework for background estimation by partitioning events into one signal-enhanced region (A) and three background-enhanced control regions (B, C, and D) via two statistically independent variables. In practice, even slight correlations between the two variables can significantly undermine the method's performance. Thus, choosing appropriate variables by hand can present a formidable challenge, especially when background and signal differ only subtly. To address this issue, the ABCD with distance correlation (ABCDisCo) method was developed to construct two artificial variables from the output scores of a neural network trained to maximize signal-background discrimination while minimizing correlations using the distance correlation measure. However, relying solely on minimizing the distance correlation can yield undesirable characteristics in the resulting distributions, which may compromise the validity of the background prediction obtained using this method. The ABCDisCo training enhanced with closure (ABCDisCoTEC) method is introduced to solve this issue by directly minimizing the nonclosure, expressed as a dedicated differentiable loss term. This extended method is applied to a data set of proton-proton collisions at a center-of-mass energy of 13 TeV recorded by the CMS detector at the CERN LHC. Additionally, given the complexity of the minimization problem with constraints on multiple loss terms … |
| Anahtar Kelimeler |
