Sample-Size Scalability and Representation Sensitivity of Quantum Machine Learning Models for Higgs Boson Event Classification
Abstract
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.
Related
- Scalability bottlenecks in entanglement-based satellite quantum key distribution: downlink asymmetry, orbit dependence and finite-key effects
- Representation Learning with Quantum Signal Processing
- Quantum Representation Learning Beyond Pairwise Fidelity
- Towards unsupervised representation learning for quantum data: quantum models with inference and generation
- Facial classification Using Hybrid Quantum Machine Learning
BibTeX
@misc{bhardwaj2026samplesize,
title = {{Sample-Size Scalability and Representation Sensitivity of Quantum Machine Learning Models for Higgs Boson Event Classification}},
author = {Shiksha Bhardwaj and Ankhi Roy},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.14533},
eprint = {2610.14533},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.14533},
}
RIS
TY - UNPB
TI - Sample-Size Scalability and Representation Sensitivity of Quantum Machine Learning Models for Higgs Boson Event Classification
AU - Shiksha Bhardwaj
AU - Ankhi Roy
PY - 2026
DA - 2026/10/10
PB - arXiv
DO - 10.48550/arxiv.2610.14533
UR - https://arxiv.org/abs/2610.14533
AB - 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.
ER -