Título: Uma Proposta para Calibração do Sistema Online de Seleção de Eventos no Detector ATLAS utilizando Gradient Boosted Decision Trees
Autores: Eduardo Simas, Paulo Farias, Edmar Souza, Juan Marin, Werner Freund, Jose Seixas, João Fonseca and Bertrand Laforge
Resumo:
Particle physics experiments deal with a huge volume of information and a complex sequential processing chain for online selection (trigger) of events. In the ATLAS experiment at the Large Hadron Collider (LHC), the trigger system is responsible for choosing the events that will be recorded on permanent media for future analysis and operates sequentially in two levels of selection. The estimated value of energy deposited by particles in the detector is an important parameter for the online selection process. In this work, a calibration method based on gradient boosted decision trees ensemble is proposed to improve the quality of the energy estimated in the second trigger stage of the ATLAS detector. With the proposed method it is possible, at the same time, to reduce computational requirements and increase the selection efficiency of electromagnetic particles (electrons).
Palavras-chave:
Particle physics, Calibration, gradient boosting, decision trees..
Páginas: 8
Código DOI: 10.21528/CBIC2021-149
Artigo em pdf: CBIC_2021_paper_149.pdf
Arquivo BibTeX: CBIC_2021_149.bib