Speaker
Description
The identification of jets coming from heavy-flavor quarks, namely bottom and charm quarks, is an important and non-trivial task at the LHC experiments. The classification of jets coming from bottom and anti-bottom quarks at the LHCb experiment would allow to perform physics measurements, such as the forward-central charge asymmetry, to constrain the Standard Model predictions and/or find possible signals of New Physics. While recently Machine Learning algorithms have played an important role in exploiting the jet substructure, there is room for improvement in the jet identification by exploiting the particles correlations. In this paper, we present a brand new approach to identify the charge of jets produced by bottom quarks, based on Quantum Machine Learning techniques. Data are embedded in a quantum circuit through a quantum feature map, a training procedure is performed, and the measurement of final state observables is mapped to a binary classification label. The models are trained and evaluated using LHCb Open Data obtained from LHCb simulations and the tagging performance is compared with the Muon Tagging algorithm used so far at LHCb, and a classical Deep Neural Network model.
