Towards a method to anticipate dark matter signals with deep learning at the LHC
EntityUAM. Departamento de Física Teórica
10.21468/SCIPOSTPHYS.12.2.063SciPost Physics 12.2 (2022): 063
Funded byThe work of EA is partially supported by the “Atracción de Talento” program (Modalidad 1) of the Comunidad de Madrid (Spain) under the grant number 2019-T1/TIC-14019 and by the Spanish Research Agency (Agencia Estatal de Investigación) through the grant IFT Centro de Excelencia Severo Ochoa SEV-2016-0597. This work has been also partially supported by CONICET and ANPCyT under projects PICT 2016-0164, PICT 2017-0802, PICT 2017-2751, PICT 2017- 2765, and PICT 2018-03682
ProjectGobierno de España. SEV-2016-0597
SubjectsDark Matter; Higgs; Scalar; Física
Rights© E. Arganda et al.
Esta obra está bajo una Licencia Creative Commons Atribución 4.0 Internacional.
We study several simplified dark matter (DM) models and their signatures at the LHC using neural networks. We focus on the usual monojet plus missing transverse energy channel, but to train the algorithms we organize the data in 2D histograms instead of event-by-event arrays. This results in a large performance boost to distinguish between standard model (SM) only and SM plus new physics signals. We use the kinematic monojet features as input data which allow us to describe families of models with a single data sample. We found that the neural network performance does not depend on the simulated number of background events if they are presented as a function of S/pB, for reasonably large B, where S and B are the number of signal and background events per histogram, respectively. This provides flexibility to the method, since testing a particular model in that case only requires knowing the new physics monojet cross section. Furthermore, we also discuss the network performance under incorrect assumptions about the true DM nature. Finally, we propose multimodel classifiers to search and identify new signals in a more general way, for the next LHC run
Google Scholar:Arganda Carreras, Ernesto - Medina, Aníbal D. - Pérez, Andrés D. - Szynkman, Alejandro
This item appears in the following Collection(s)
Showing items related by title, author, creator and subject.
A method for approximating optimal statistical significances with machine-learned likelihoods Arganda Carreras, Ernesto; Marcano Imaz, Xabier; Martín Lozano, Víctor; Medina, Aníbal D.; Pérez, Andrés D.; Szewc, Manuel; Szynkman, Alejandro
Model-independent search strategy for the lepton-flavor-violating heavy Higgs boson decay to τμ at the LHC Arganda, Ernesto; Marcano, Xabier; Mileo, Nicolás I.; Morales, Roberto A.; Szynkman, Alejandro
Higgs effective Hlilj vertex from heavy νR and applications to LFV Herrero, María J.; Arganda, Ernesto; Marcano, Xavier; Morales, Roberto; Szynkman, Alejandro