Speaker
Description
The LHC detectors are going to be upgraded to profit from the substantial increase of the LHC instantaneous luminosity. In particular, in the forthcoming High Luminosity phase of ATLAS and CMS the number of proton–proton interactions per beam crossing is expected to rise by a factor three, averaging 140–200 in future Runs 4 and 5. Similar upgrades are also planned for the LHCb and ALICE detectors in Run5. This increase in luminosity leads to larger, denser events, and, consequently, greater complexity in reconstructing charged particle tracks, thus motivating frontier research in new technologies.
Quantum computing and machine learning methods have proved themselves to be two of the most promising emerging computing technologies of the last years. At the intersection of these two fields, we upgrade and characterize a quantum graph neural network (QGNN) architecture for charged particle track reconstruction, evaluated on a simulated high luminosity dataset.
The model operates on a set of event graphs, each built from the hits generated in tracking detector layers by particles produced in proton collisions, performing a classification of the possible hit connections between adjacent layers. In this approach, the QGNN is designed as a hybrid architecture, interleaving classical feedforward networks with parametrized quantum circuits. We characterize the interplay between the classical and quantum components. We report on the principal upgrades to the original design, and present new evidence of improved training behavior, specifically in terms of convergence toward the final trained configuration. Finally, we provide an outlook on noisy-intermediate-scale QML methods for high energy physics.
| Sessions | Quantum Machine Learning: |
|---|---|
| Invited | No |