Indexado en

Licencia y uso

Citaciones

Cited 4 times in

Cited 4 times in

Altmetrics

Grant support

We thank Xabier Marcano, Manuel Szewc, Alejandro Szynkman, and Hernan Wahlberg for fruitful discussions on the implementation of KDE in the MLL method. This work is partially supported by the "Atraccion de Talento" program (Modalidad 1) of the Comunidad de Madrid (Spain) under the Grant number 2019-T1/TIC-14019 (EA, RMSS), by the Spanish Research Agency (Agencia Estatal de Investigacion) through the Grant IFT Centro de Excelencia Severo Ochoa No CEX2020-001007-S (EA, MdlR, AP, RMSS) and the Grant PID2021-124704NB-I00 (EA, RMSS), funded by MCIN/AEI/10.13039/501100011033, and by CONICET and ANPCyT under projects PICT 2017-2751 and PICT 2018-03682 (EA, AP). MdlR also acknowledges financial support from the ComunidadAutonoma de Madrid through the Grant SI2/PBG/2020-00005.

Impacto en los Objetivos de Desarrollo Sostenible (ODS)

Análisis de autorías institucional

Compartir

Machine-learned exclusion limits without binning

Publicado en:EUROPEAN PHYSICAL JOURNAL C. 83 (12): 1158- - 2023-12-19 83(12), DOI: 10.1140/epjc/s10052-023-12314-z

Autores: Arganda, Ernesto; Perez, Andres D; de los Rios, Martin; Sanda Seoane, Rosa Maria

Afiliaciones

Resumen

Machine-learned likelihoods (MLL) combines machine-learning classification techniques with likelihoodbased inference tests to estimate the experimental sensitivity of high-dimensional data sets. We extend theMLLmethod by including kernel density estimators (KDE) to avoid binning the classifier output to extract the resulting one-dimensional signal and background probability density functions. We first test our method on toy models generated with multivariate Gaussian distributions, where the true probability distribution functions are known. Later, we apply the method to two cases of interest at the LHC: a search for exotic Higgs bosons, and a Z' boson decaying into lepton pairs. In contrast to physical-based quantities, the typical fluctuations of the ML outputs give non-smooth probability distributions for puresignal and pure-background samples. The non-smoothness is propagated into the density estimation due to the good performance and flexibility of the KDE method. We study its impact on the final significance computation, and we compare the results using the average of several independent ML output realizations, which allows us to obtain smoother distributions. We conclude that the significance estimation turns out to be not sensible to this issue.

Palabras clave