Browsing by Author "Zavala Torres, Javier Alejandro"

Now showing 1 - 1 of 1

Identification of boosted tau pairs at atlas experiment using deep neural networks
(Universidad Técnica Federico Santa María, 2022-01) Zavala Torres, Javier Alejandro; Torres Lopez, Claudio Esteban; Universidad Técnica Federico Santa María. Departamento de Informática; Pezoa Rivera, Raquel Andrea
Measurements related to high-energy physics events producing a pair of tau particles are essential to search for new physics because it is one of the most important ecays of the Higgs boson. Scientists in CERN’s ATLAS experiment are working on the development of algorithms dedicated to the identifica􀆟on of pairs of tau particles. This work focuses on the development of a computational algorithm for identifying the generation of a pair of tau particles of the type boosted, which includes (i) reduc􀆟on and validation of the data set composed of the pair signal of taus boosted and the background corresponding to hadron jets (ii) development of an algorithm based on deep neural networks to classify events into signal or background, (iii) validation of the proposed method and comparison with the current solution. The experiments carried out resulted in an F1 score of 0.989, reaching 99.9 % of the current solution, and a background rejection of 9, 460 · 102 was obtained for the working point very loose with signal efficiency 0.95, reaching 74.6 % of the current solution.