CERN Accelerating science

Article
Report number	arXiv:2212.07274
Title	Robust Neural Particle Identification Models
Author(s)	Ryzhikov, Artem (Higher Sch. of Economics, Moscow) ; Temirkhanov, Aziz (Higher Sch. of Economics, Moscow) ; Derkach, Denis (Higher Sch. of Economics, Moscow) ; Hushchyn, Mikhail (Higher Sch. of Economics, Moscow) ; Kazeev, Nikita (Higher Sch. of Economics, Moscow) ; Mokhnenko, Sergei (Higher Sch. of Economics, Moscow)
Collaboration	LHCb Collaboration
Publication	2023
Imprint	2022-12-14
Number of pages	5
Note	proceedings of ACAT-2021
In:	J. Phys. : Conf. Ser. 2438, 1 (2023) pp.012119
In:	20th International Workshop on Advanced Computing and Analysis Techniques in Physics Research (ACAT 2021), Daejeon, Korea, 29 Nov - 3 Dec 2021, pp.012119
DOI	10.1088/1742-6596/2438/1/012119
Subject category	physics.ins-det ; Detectors and Experimental Techniques ; hep-ex ; Particle Physics - Experiment
Accelerator/Facility, Experiment	CERN LHC ; LHCb
Abstract	The volume of data processed by the Large Hadron Collider experiments demands sophisticated selection rules typically based on machine learning algorithms. One of the shortcomings of these approaches is their profound sensitivity to the biases in training samples. In the case of particle identification (PID), this might lead to degradation of the efficiency for some decays not present in the training dataset due to differences in input kinematic distributions. In this talk, we propose a method based on the Common Specific Decomposition that takes into account individual decays and possible misshapes in the training data by disentangling common and decay specific components of the input feature set. We show that the proposed approach reduces the rate of efficiency degradation for the PID algorithms for the decays reconstructed in the LHCb detector.
Copyright/License	publication: (License: CC-BY-3.0) preprint: (License: CC BY 4.0)

Corresponding record in: Inspire