Unravelling physics beyond the standard model with classical and quantum anomaly detection

Abstract

Much hope for finding new physics phenomena at microscopic scale relies on the observations obtained from High Energy Physics experiments, like the ones performed at the Large Hadron Collider (LHC). However, current experiments do not indicate clear signs of new physics that could guide the development of additional Beyond Standard Model (BSM) theories. Identifying signatures of new physics out of the enormous amount of data produced at the LHC falls into the class of anomaly detection and constitutes one of the greatest computational challenges. In this article, we propose a novel strategy to perform anomaly detection in a supervised learning setting, based on the artificial creation of anomalies through a random process. For the resulting supervised learning problem, we successfully apply classical and quantum Support Vector Classifiers (CSVC and QSVC respectively) to identify the artificial anomalies among the SM events. Even more promising, we find that employing an SVC trained to identify the artificial anomalies, it is possible to identify realistic BSM events with high accuracy. In parallel, we also explore the potential of quantum algorithms for improving the classification accuracy and provide plausible conditions for the best exploitation of this novel computational paradigm.

Optimizing Quantum Classification Algorithms on Classical Benchmark Datasets

“…Once such a quantum embedding is realized, a decision rule to carry out the desired classification task can be obtained directly from the fidelity between the encoded feature vectors [ 20 ] or by passing the resulting quantum kernel to a classical support vector machine [ 18 , 19 ]. Soon after the introduction of quantum kernel methods, it was shown that quantum kernel methods, equipped with the right quantum feature maps, can solve certain specifically designed problems more efficiently than any known classical counterpart [ 21 ], thus motivating a large body of research aimed at finding similar advantages in more generic and applied contexts [ 22 , 23 , 24 ], including for the following: data analysis for high-energy physics [ 25 , 26 , 27 ], quantum phase classification [ 28 ], fraud detection [ 29 ] and virtual screening for drug discovery [ 30 ]. While some promising examples were identified [ 27 , 30 ], only proof-of-principle results have been achieved so far, mostly based on empirical considerations.…”