We report on a consistent comparison between techniques of quantum and classical machine learning applied to the classification of signal and background events for the Vector Boson Scattering processes, studied at the Large Hadron Collider installed at the CERN laboratory. Quantum machine learning algorithms based on variational quantum circuits are run on freely available quantum computing hardware, showing very good performances as compared to deep neural networks run on classical computing facilities. In particular, we show that such kind of quantum neural networks is able to correctly classify the targeted signal with an Area Under the characteristic Curve (AUC) that is very close to the one obtained with the corresponding classical neural network, but employing a much lower number of resources, as well as less variable data in the training set. Albeit giving a proof-of-principle demonstration with limited quantum computing resources, this work represents one of the first steps towards the use of near term and noisy quantum hardware for practical event classification in High Energy Physics experiments.

Cugini, D., Gerace, D., Govoni, P., Perego, A., Valsecchi, D. (2023). Comparing quantum and classical machine learning for Vector Boson Scattering background reduction at the Large Hadron Collider. QUANTUM MACHINE INTELLIGENCE, 5(2) [10.1007/s42484-023-00106-3].

Comparing quantum and classical machine learning for Vector Boson Scattering background reduction at the Large Hadron Collider

Cugini D.;Govoni P.;Perego A.;Valsecchi D.
2023

Abstract

We report on a consistent comparison between techniques of quantum and classical machine learning applied to the classification of signal and background events for the Vector Boson Scattering processes, studied at the Large Hadron Collider installed at the CERN laboratory. Quantum machine learning algorithms based on variational quantum circuits are run on freely available quantum computing hardware, showing very good performances as compared to deep neural networks run on classical computing facilities. In particular, we show that such kind of quantum neural networks is able to correctly classify the targeted signal with an Area Under the characteristic Curve (AUC) that is very close to the one obtained with the corresponding classical neural network, but employing a much lower number of resources, as well as less variable data in the training set. Albeit giving a proof-of-principle demonstration with limited quantum computing resources, this work represents one of the first steps towards the use of near term and noisy quantum hardware for practical event classification in High Energy Physics experiments.
Articolo in rivista - Articolo scientifico
Large Hadron Collider; Quantum computing; Quantum machine learning; Vector Boson Scattering; Variational quantum circuits;
English
9-ago-2023
2023
5
2
35
none
Cugini, D., Gerace, D., Govoni, P., Perego, A., Valsecchi, D. (2023). Comparing quantum and classical machine learning for Vector Boson Scattering background reduction at the Large Hadron Collider. QUANTUM MACHINE INTELLIGENCE, 5(2) [10.1007/s42484-023-00106-3].
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/476044
Citazioni
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 1
Social impact