Robust machine-learning methods are becoming increasingly important for high-energy physics data analysis as experiments enter the era of higher luminosity and future higher-energy colliders. Detector degradation, changing running conditions and calibration drift can shift data distributions, causing models trained on clean reference samples to degrade after deployment. We investigate whether parameterised quantum models provide a useful inductive bias for robust collider-event selection in two complementary settings. In the unsupervised study, quantum autoencoders trained on background events are compared with classical and variational autoencoders for anomaly detection. In the supervised study, quantum classifiers with data reuploading are trained to distinguish a supersymmetric signal from background and are compared with linear and multilayer-perceptron classifiers. All models are trained under reference conditions and subsequently evaluated under controlled feature-level smearing while their parameters and preprocessing transformations are held fixed. On clean inputs, the quantum autoencoders achieve competitive anomaly-detection performance, including in the low-false-positive-rate regime relevant for triggering, while the deeper data-reuploading classifier attains discrimination comparable to the non-linear classical baseline. Under smearing, the quantum models generally exhibit smaller shifts in their output scores and retain their discrimination more effectively than the expressive classical baselines. These results suggest that parameterised quantum models can provide a useful robustness inductive bias for collider-event selection and motivate further studies with realistic detector systematics, finite-shot statistics and quantum-device noise.
Robust machine-learning methods are becoming increasingly important for high-energy physics data analysis as experiments enter the era of higher luminosity and future higher-energy colliders. Detector degradation, changing running conditions and calibration drift can shift data distributions, causing models trained on clean reference samples to degrade after deployment. We investigate whether parameterised quantum models provide a useful inductive bias for robust collider-event selection in two complementary settings. In the unsupervised study, quantum autoencoders trained on background events are compared with classical and variational autoencoders for anomaly detection. In the supervised study, quantum classifiers with data reuploading are trained to distinguish a supersymmetric signal from background and are compared with linear and multilayer-perceptron classifiers. All models are trained under reference conditions and subsequently evaluated under controlled feature-level smearing while their parameters and preprocessing transformations are held fixed. On clean inputs, the quantum autoencoders achieve competitive anomaly-detection performance, including in the low-false-positive-rate regime relevant for triggering, while the deeper data-reuploading classifier attains discrimination comparable to the non-linear classical baseline. Under smearing, the quantum models generally exhibit smaller shifts in their output scores and retain their discrimination more effectively than the expressive classical baselines. These results suggest that parameterised quantum models can provide a useful robustness inductive bias for collider-event selection and motivate further studies with realistic detector systematics, finite-shot statistics and quantum-device noise.
The lack of evidence for new interactions and particles at the Large Hadron Collider (LHC) has motivated the high-energy physics community to explore model-agnostic data-analysis approaches to search for new physics. Autoencoders are unsupervised machine learning models based on artificial neural networks, capable of learning background distributions. We study quantum autoencoders based on variational quantum circuits for the problem of anomaly detection at the LHC. For a QCD tt background and resonant heavy-Higgs signals, we find that a simple quantum autoencoder outperforms classical autoencoders for the same inputs and trains very efficiently. Moreover, this performance is reproducible on present quantum devices. This shows that quantum autoencoders are good candidates for analysing high-energy physics data in future LHC runs.
Anomaly detection in high-energy physics is essential for identifying rare events that may signal physics beyond the Standard Model. While classical autoencoders are widely used, their latent spaces lack quantum-specific structure such as entanglement. To address this, we propose a hybrid Quantum Variational Autoencoder (QVAE) that incorporates a parameterized quantum circuit to encode latent variables and extract entanglement-based features. Using the LHC Olympics anomaly detection dataset, this research measures single-qubit von Neumann and Rényi-2 entropies from reduced density matrices of the latent quantum state. Experiments show that anomalous events exhibit statistically distinguishable entanglement signatures compared to background, confirmed by Kolmogorov-Smirnov tests (p < 0.003). Entanglement-based anomaly scores achieve competitive ROC-AUC performance and improve further when combined with reconstruction error. Ablation results demonstrate that removing entangling gates eliminates this separation, indicating that entanglement contributes meaningfully to anomaly detection. The proposed QVAE provides a promising direction for interpretable quantum-enhanced analysis of collider data.
Quantum machine learning (QML) algorithms in high energy physics (HEP) can efficiently represent and leverage long-range, high-order correlations in high-dimensional collider data, potentially with fewer parameters and favorable scaling relative to classical models. Deployment of QML in real-time collider applications such as trigger systems requires the ability to emulate and compile quantum circuits classically, then synthesize the resulting quantum gates onto low-latency hardware accelerators, namely field-programmable gate arrays (FPGAs). We present a study of variational quantum autoencoder models for real-time anomaly detection triggers in modern collider experiments. The models achieve performance comparable to state-of-the-art classical approaches and, after FPGA synthesis, satisfy resource usage and timing constraints consistent with trigger applications in future colliders. This work provides one of the first FPGA implementations of QML models for HEP triggers, enabling higher-capability models in today's classical data acquisition pipelines while advancing quantum readiness of collider experiment infrastructure.
Quantum machine learning offers new approaches to multivariate event classification in high-energy physics, but its performance relative to optimized classical methods remains uncertain under limited training data and restricted quantum resources. We investigate the sample-size scalability and representation sensitivity of a quantum support vector machine and a variational quantum classifier for Higgs-boson signal--background discrimination using simulated events from the ATLAS Higgs Machine Learning Challenge. The models are trained on balanced subsets ranging from 250 to 4000 events and evaluated on held-out samples preserving the original class distribution. Principal component analysis (PCA) and autoencoder (AE) representations are examined to reduce the input dimensionality and quantum-resource requirements. Four optimized classical classifiers are used as benchmarks under a common preprocessing, validation, and evaluation protocol. We assess training-sample scalability in classical and quantum classifiers using a four-component PCA representation and examine representation sensitivity through a secondary matched-event comparison of four- and eight-component PCA representations. Within the evaluated Higgs-boson classification setting, the QSVM showed stronger performance than the VQC at larger training sample sizes, while remaining below the best-performing classical classifiers.
Anomaly detection (AD) underpins fraud prevention, network security, industrial condition monitoring, medical diagnosis, and searches for new physics. However, it remains challenging due to rare anomalies, scarce labeled data, and the high dimensionality and strong correlation of sensor data. Many classical machine learning (ML) algorithms are designed to address AD in various settings, but the growth of quantum computing opens a new path for these problems. Quantum machine learning (QML) offers new tools, from exploring exponentially large Hilbert spaces and promising speed-ups to the native processing of quantum data. Thus, this review connects the two fields. We first present the basics of quantum computing, quantum algorithms, and QML with the most influential literature, and formalize the classical anomaly scores that quantum methods inherit. We then systematically review 30 primary studies of quantum anomaly detection (QAD) published between 2018 and 2026, express each method family through its anomaly score, and organize the field into a taxonomy of six families: quantum kernel methods, quantum autoencoders, quantum generative models, variational quantum classifiers, quantum linear-algebra methods, and detectors for quantum data. A cross-study analysis shows that the field has moved from asymptotic speed-up proposals to small hybrid experiments, that kernels and autoencoders dominate, that only nine studies run on gate-based hardware, and that none yet demonstrates a robust, practically relevant advantage over well-tuned classical detectors. We close with the main obstacles, such as data embedding, barren plateaus, kernel concentration, and weak baselines, and a research roadmap that prioritizes quantum data, fair benchmarks against classical baselines, and robust detectors for sequential and distributed data.
A comparative study of variational quantum classifiers and classical machine learning for tW−ttˉ discrimination at the LHC
Precise measurements of single top-quark production in the associated tW-channel require efficient discrimination against the dominant ttˉ background. In this work, we investigate the applicability of quantum machine learning to this classification problem using the Variational Quantum Classifier (VQC) algorithm. Simulated proton-proton collision events from CMS Open Data corresponding to Run-2 are employed to train and evaluate the VQC model. Its performance is compared with two established classical approaches, Boosted Decision Trees (BDTs) and Multi-Layer Perceptron (MLP) using both machine learning metrics and physics-motivated metrics that describe the Signal Significance as a function of Signal Efficiency. The comparison is performed separately for the dileptonic e−μ channel and for the resolved and boosted hadronic topologies. In addition, signal-enriched and background-enriched phase-space regions are studied to assess the dependence of classifier performance on event topology. Our results provide a systematic benchmark for the application of VQC for a realistic High Energy Physics classification task and identify the analysis regimes wherein quantum-inspired models offer competitive discrimination with respect to established classical techniques.