LIGHTWEIGHT COMPARATIVE EVALUATION OF QUANTUM AND HYBRID QUANTUM-CLASSICAL MACHINE LEARNING MODELS FOR INTRUSION DETECTION
DOI:
https://doi.org/10.68302/std2026.vol1.207Keywords:
Cybersecurity, hybrid quantum-classical models, intrusion detection, quantum machine learningAbstract
Quantum machine learning (QML) has recently emerged as a promising research direction at the intersection of artificial intelligence, quantum computing, and cybersecurity. Among its potential applications, intrusion detection remains particularly relevant due to the growing complexity of cyber threats and the increasing need for adaptive and efficient detection mechanisms. This paper explores the applicability of lightweight quantum and hybrid quantum-classical machine learning approaches for intrusion detection in resource-constrained experimental settings. The study is designed as a comparative analysis of selected models, including variational and kernel-based quantum techniques, alongside a conventional classical baseline, in order to evaluate their practical potential for cybersecurity tasks.
The experimental framework relies on a reduced and preprocessed cybersecurity dataset and is implemented in a simulated quantum environment using accessible open-source tools. The comparison considers standard performance indicators such as classification effectiveness and computational efficiency, while also reflecting on broader dimensions of model suitability for security-oriented applications. In addition to the technical evaluation, the paper addresses several ethical considerations associated with the use of QML in cybersecurity, including limited transparency, risks related to biased or imbalanced training data, and the operational implications of incorrect classifications in intrusion detection scenarios.
Rather than claiming immediate quantum advantage, the study aims to provide a realistic and methodologically grounded perspective on the current role of QML in cybersecurity research. The expected contribution is twofold - first, to outline a feasible experimental approach for evaluating QML methods in intrusion detection and second, to highlight the importance of combining performance-oriented assessment with ethical and interpretability-related considerations when examining emerging intelligent security solutions.
Supporting Agencies
ACKNOWLEDGEMENTS This article is the result of research funded by the National Science Fund of Bulgaria under project No. BG-175467353-2024-10-0013, contract No. КП-06-М87/5, dated December 6, 2024. The project is titled Artificial Intelligence and Cybersecurity: Navigating the Ethical Challenges.Downloads
References
[1] A. L. Buczak and E. Guven, “A Survey of Data Mining and Machine Learning Methods for Cyber Security Intrusion Detection,” IEEE Communications Surveys and Tutorials, vol. 18, no. 2, pp. 1153–1176, Apr. 2016, doi: 10.1109/COMST.2015.2494502;PAGE:STRING:ARTICLE/CHAPTER.
[2] R. Sommer and V. Paxson, “Outside the Closed World: On Using Machine Learning For Network Intrusion Detection”.
[3] “NSL-KDD.” Accessed: May 10, 2026. [Online]. Available: https://www.kaggle.com/datasets/hassan06/nslkdd
[4] “The UNSW-NB15 Dataset | UNSW Research.” Accessed: May 10, 2026. [Online]. Available: https://research.unsw.edu.au/projects/unsw-nb15-dataset
[5] J. Biamonte, P. Wittek, N. Pancotti, P. Rebentrost, N. Wiebe, and S. Lloyd, “Quantum machine learning,” Nature 2017 549:7671, vol. 549, no. 7671, pp. 195–202, Sep. 2017, doi: 10.1038/nature23474.
[6] J. Preskill, “Quantum Computing in the NISQ era and beyond,” Quantum, vol. 2, p. 79, Aug. 2018, doi: 10.22331/q-2018-08-06-79.
[7] M. Cerezo et al., “Variational quantum algorithms,” Nature Reviews Physics 2021 3:9, vol. 3, no. 9, pp. 625–644, Aug. 2021, doi: 10.1038/s42254-021-00348-9.
[8] M. Schuld and N. Killoran, “Quantum Machine Learning in Feature Hilbert Spaces,” Phys. Rev. Lett., vol. 122, no. 4, p. 040504, Feb. 2019, doi: 10.1103/PhysRevLett.122.040504.
[9] V. Havlíček et al., “Supervised learning with quantum-enhanced feature spaces,” Nature 2019 567:7747, vol. 567, no. 7747, pp. 209–212, Mar. 2019, doi: 10.1038/s41586-019-0980-2.
[10] F. Pedregosa FABIANPEDREGOSA et al., “Scikit-learn: Machine Learning in Python,” Journal of Machine Learning Research, vol. 12, no. 85, pp. 2825–2830, 2011, Accessed: May 10, 2026. [Online]. Available: http://jmlr.org/papers/v12/pedregosa11a.html
[11] G. Aleksandrowicz et al., “Qiskit: An Open-source Framework for Quantum Computing”, doi: 10.5281/ZENODO.2562111.
[12] “GitHub - qiskit-community/qiskit-machine-learning: An open-source library built on Qiskit for quantum machine learning tasks at scale on quantum hardware and classical simulators · GitHub.” Accessed: May 10, 2026. [Online]. sAvailable: https://github.com/qiskit-community/qiskit-machine-learning
[13] N. Institute of Standards, “Artificial Intelligence Risk Management Framework (AI RMF 1.0),” 2020, doi: 10.6028/NIST.AI.100-1.
[14] A. J. Smola and B. Schölkopf, “A tutorial on support vector regression,” Stat. Comput., vol. 14, no. 3, pp. 199–222, Aug. 2004, doi: 10.1023/B:STCO.0000035301.49549.88.
[15] “QSVC - Qiskit Machine Learning 0.9.0.” Accessed: May 10, 2026. [Online]. Available: https://qiskit-community.github.io/qiskit-machine-learning/stubs/qiskit_machine_learning.algorithms.QSVC.html
[16] “VQC - Qiskit Machine Learning 0.9.0.” Accessed: May 10, 2026. [Online]. Available: https://qiskit-community.github.io/qiskit-machine-learning/stubs/qiskit_machine_learning.algorithms.VQC.html
[17] “Qiskit Aer 0.17.1.” Accessed: May 10, 2026. [Online]. Available: https://qiskit.github.io/qiskit-aer/
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Genadiy Gospodinov, Tamara Ristovska, Tito Titov, Kiril Nikolov, Andriyan Stoilov, Kliment Toshkov
This work is licensed under a Creative Commons Attribution 4.0 International License.
