Quantum anomaly detection for collider physics

Abstract

We explore the use of Quantum Machine Learning (QML) for anomaly detection at the Large Hadron Collider (LHC). In particular, we explore a semi-supervised approach in the four-lepton final state where simulations are reliable enough for a direct background prediction. This is a representative task where classification needs to be performed using small training datasets — a regime that has been suggested for a quantum advantage. We find that Classical Machine Learning (CML) benchmarks outperform standard QML algorithms and are able to automatically identify the presence of anomalous events injected into otherwise background-only datasets.

Generative invertible quantum neural networks

“…This theoretical groundwork is bolstered by several instances of specifically constructed QML circuits presenting significantly more efficient solutions than classically possible to specially designed problems [40][41][42][43]. QNNs have already been successfully applied to relatively limited high-energy physics problems [21,25,[44][45][46][46][47][48][49][50][51], along non-QML approaches [52][53][54][55][56]. However, to our knowledge, there has not yet been an attempt to construct an invertible QNN that can be used as a density estimator through its invertibility for generative tasks.…”

Guided quantum compression for high dimensional data classification

“…Recent studies have highlighted guarantees regarding the expressivity, generalisation power, and trainability of quantum models [25][26][27][28][29][30]. Moreover, the efficacy of applying QML models to High Energy Physics (HEP) data analysis is exemplified in studies for classification [31][32][33][34][35][36], reconstruction [37][38][39], anomaly detection [40][41][42][43], and Monte Carlo integration [44,45]. A summary of advancements in QML applied to HEP is found in [46].…”

Generative Invertible Quantum Neural Networks

“…This theoretical groundwork is bolstered by several instances of specifically constructed QML circuits presenting significantly more efficient solutions than classically possible to specially designed problems [40][41][42][43]. QNNs have already been successfully applied to relatively limited high-energy physics problems [21,25,[44][45][46][46][47][48], along non-QML approaches [49][50][51][52][53]. However, to our knowledge, there has not yet been an attempt to construct an invertible QNN that can be used as a density estimator through its invertibility for generative tasks.…”