https://doi.org/10.1140/epjc/s10052-020-8082-8
Regular Article - Theoretical Physics
Jet tagging made easy
1
Instituto de Física Teórica UAM-CSIC, Campus de Cantoblanco, 28049, Madrid, Spain
2
Departamento de Física Teórica, Universidad Autónoma de Madrid, Cantoblanco, 28049, Madrid, Spain
* e-mail: jaas@ugr.es
Received:
17
March
2020
Accepted:
23
May
2020
Published online:
13
June
2020
We develop taggers for multi-pronged jets that are simple functions of jet substructure (so-called ‘subjettiness’) variables. These taggers can be approximately decorrelated from the jet mass in a quite simple way. Specifically, we use a Logistic Regression Design (LoRD) which, even being one of the simplest machine learning classifiers, shows a performance which surpasses that of simple variables used by the ATLAS and CMS Collaborations and is not far from more complex models based on neural networks. Contrary to the latter, our method allows for an easy implementation of tagging tasks by providing a simple and interpretable analytical formula with already optimised parameters.
© The Author(s), 2020