Towards Quantum-Enhanced Machine Learning for Network Intrusion Detection
The DOI matches its Crossref record; 34 of the 34 references it cites were found in a registry or an index.
Abstract
Network Intrusion Detection Systems (NIDSs) are commonly used today to detect malicious activities. Quantum computers, despite not being practical yet, are becoming available for experimental purposes. We present the first approach for applying unsupervised Quantum Machine Learning (QML) in the context of network intrusion detection from the perspective of quantum information, based on the concept of quantum-assisted ML. We evaluate it using IBM QX in simulation mode and show that the accuracy of a Quantum-Assisted NIDS, based on our approach, can be high, rivaling with the the best conventional SVM results, with a dependence on the characteristics of the dataset.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“For example, [17] reviews a quantum support vector machine (QSVM) constructed to detect DDoS attacks. [18] presents the experimental study of the QSVM running with conventional NSL-KDD and NB-15 datasets, that denotes the QSVM’s readiness to the intrusion detection on common attack datasets.”
From Security intrusion detection using quantum machine learning techniques (2022) · cites this paper as [18]
Citations
Cites
Quantum Computing in the NISQ era and beyond
Preskill, J. (2018) · Quantum · doi:10.22331/q-2018-08-06-79 · Cited by 8,830 · cited 2×
“QSVMs are an interesting application for NISQ computers [22], [23].”
BackgroundMethodQuantum supremacy using a programmable superconducting processor
Arute, F., Arya, K., Babbush, R., Bacon, D., Bardin, J. C., Barends, R., Biswas, R., Boixo, S., Brandao, F. G. S. L., Buell, D. A., et al. (2019) · Nature · doi:10.1038/s41586-019-1666-5 · Cited by 7,720 · cited 1×
“A definitive proof of quantum dominance is predicted in the near future [9]–[11], and one argument has already been made that it has been accomplished [12], although it has been disputed by competitors [13].”
BackgroundA detailed analysis of the KDD CUP 99 data set
Tavallaee, M., Bagheri, E., Lu, W., & Ghorbani, A. A. (2009) · 2009 IEEE Symposium on Computational Intelligence for Security and Defense Applications · doi:10.1109/cisda.2009.5356528 · Cited by 5,396
Moustafa, N., & Slay, J. (2015) · 2015 Military Communications and Information Systems Conference (MilCIS) · doi:10.1109/milcis.2015.7348942 · Cited by 4,033 · cited 4×
“The phases presented in Section IV were executed the following way: 1) Obtaining the input data: the flows were extracted from the KDD-NSL and NB15 datasets: 3) Encoding of the data: we considered an implementation based on the autoencode() function provided by the Ruta package [31], [32] for the R statistical computin…”
BackgroundInfluentialSupervised learning with quantum-enhanced feature spaces
Havlíček, V., Córcoles, A. D., Temme, K., Harrow, A. W., Kandala, A., Chow, J. M., & Gambetta, J. M. (2019) · Nature · doi:10.1038/s41586-019-0980-2 · Cited by 2,572
Anomaly-based network intrusion detection: Techniques, systems and challenges
García-Teodoro, P., Díaz-Verdejo, J., Maciá-Fernández, G., & Vázquez, E. (2009) · Computers & Security · doi:10.1016/j.cose.2008.08.003 · Cited by 2,085 · cited 1×
“Network IDSs (NIDSs) in particular detect malicious activity in networks and are one of the best known contexts of ML application in the field [3].”
BackgroundCover, T. M. (1965) · IEEE Transactions on Electronic Computers · doi:10.1109/pgec.1965.264137 · Cited by 2,044
Quantum Support Vector Machine for Big Data Classification
Rebentrost, P., Mohseni, M., & Lloyd, S. (2014) · Physical Review Letters · doi:10.1103/physrevlett.113.130503 · Cited by 1,951
Quantum Machine Learning in Feature Hilbert Spaces
Schuld, M., & Killoran, N. (2019) · Physical Review Letters · doi:10.1103/physrevlett.122.040504 · Cited by 1,840 · cited 1×
“2) According to [23], [34] , classifiers based on quantum circuits should only provide a quantum advantage if the kernel used in the quantum space is difficult to estimate classically: K(~x, ~z) = | 〈Φ(~x) |Φ(~z) 〉 |(2) .”
BackgroundToward developing a systematic approach to generate benchmark datasets for intrusion detection
Shiravi, A., Shiravi, H., Tavallaee, M., & Ghorbani, A. A. (2012) · Computers & Security · doi:10.1016/j.cose.2011.12.012 · Cited by 1,345 · cited 3×
“• NSL-KDD: For the NSL-KDD dataset autoencode() has been tested with four activation functions – Linear, Hyperbolic Tangent, Relu and Sigmoid –, in order to understand the performance in the four cases.”
BackgroundMoustafa, N., & Slay, J. (2016) · Information Security Journal: A Global Perspective · doi:10.1080/19393555.2015.1125974 · Cited by 1,100 · cited 1×
“We designate them NSL-KDD [30] and NB15 [5], [6].”
BackgroundA survey of intrusion detection techniques for cyber-physical systems
Mitchell, R., & Chen, I.-R. (2014) · ACM Computing Surveys · doi:10.1145/2542049 · Cited by 768
Supervised Learning with Quantum Computers
Schuld, M., & Petruccione, F. (2018) · Quantum Science and Technology · doi:10.1007/978-3-319-96424-9 · Cited by 730 · cited 1×
“What is used are layers of rotation gates separated by entangling layers composed of controlled Pauli gates [35].”
MethodBuilding logical qubits in a superconducting quantum computing system
Gambetta, J. M., Chow, J. M., & Steffen, M. (2017) · npj Quantum Information · doi:10.1038/s41534-016-0004-0 · Cited by 622 · cited 1×
“On May 2016, IBM announced the first Near Term Quantum Computer, a NISQ, to be publicly available on the cloud [14].”
BackgroundEffective Intrusion Detection System Using XGBoost
Dhaliwal, S. S., Nahid, A.-A., & Abbas, R. (2018) · Information · doi:10.3390/info9070149 · Cited by 503
An Efficient Methodology for Mapping Quantum Circuits to the IBM QX Architectures
Zulehner, A., Paler, A., & Wille, R. (2019) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2018.2846658 · Cited by 372 · cited 1×
“IBM QX has been used in many experiences, e.g., quantum oscillator synchronization [15], variational quantum algorithms [16], and quantum search [17], and optimization approaches to convert common quantum circuits descriptions into the IBM QX have been proposed [18].”
MethodA revised taxonomy for intrusion-detection systems
Debar, H., Dacier, M., & Wespi, A. (2000) · Annales Des Télécommunications · doi:10.1007/bf02994844 · Cited by 359 · cited 2×
“The detection of security violations using machine learning (ML) has been extensively investigated [1]–[8].”
BackgroundVariational quantum algorithms for nonlinear problems
Lubasch, M., Joo, J., Moinier, P., Kiffner, M., & Jaksch, D. (2020) · Physical Review A · doi:10.1103/physreva.101.010301 · Cited by 332 · cited 1×
“, quantum oscillator synchronization [15], variational quantum algorithms [16], and quantum search [17], and optimization approaches to convert common quantum circuits descriptions into the IBM QX have been proposed [18].”
MethodCharte, D., Charte, F., García, S., del Jesus, M. J., & Herrera, F. (2018) · Information Fusion · doi:10.1016/j.inffus.2017.12.007 · Cited by 273 · cited 1×
“…were executed the following way: 1) Obtaining the input data: the flows were extracted from the KDD-NSL and NB15 datasets: 3) Encoding of the data: we considered an implementation based on the autoencode() function provided by the Ruta package [31], [32] for the R statistical computing package [33].”
MethodDimensionality reduction using Principal Component Analysis for network intrusion detection
Keerthi Vasan, K., & Surendiran, B. (2016) · Perspectives in Science · doi:10.1016/j.pisc.2016.05.010 · Cited by 194 · cited 1×
“Principal component analysis (PCA) is arguably the most used technique for dimensionality reduction, so it has already been used in the context of intrusion detection [19], [20].”
MethodBoth Toffoli and Controlled-NOT need little help to universal quantum computing
Shi, Y.-Y. (2003) · Quantum Information and Computation · doi:10.26421/qic3.1-7 · Cited by 193
Demonstration of quantum advantage in machine learning
Ristè, D., da Silva, M. P., Ryan, C. A., Cross, A. W., Córcoles, A. D., Smolin, J. A., Gambetta, J. M., Chow, J. M., & Johnson, B. R. (2017) · npj Quantum Information · doi:10.1038/s41534-017-0017-3 · Cited by 182 · cited 2×
“A definitive proof of quantum dominance is predicted in the near future [9]–[11], and one argument has already been made that it has been accomplished [12], although it has been disputed by competitors [13].”
BackgroundEstablishing the quantum supremacy frontier with a 281 Pflop/s simulation
Villalonga, B., Lyakh, D., Boixo, S., Neven, H., Humble, T. S., Biswas, R., Rieffel, E. G., Ho, A., & Mandrà, S. (2020) · Quantum Science and Technology · doi:10.1088/2058-9565/ab7eeb · Cited by 124 · cited 2×
“Quantum computers , and among them Noisy Intermediate-Scale Quantum (NISQ) computers , aim to leverage quantum physics to perform computational tasks beyond the capabilities of the most powerful classical computers, potentially achieving quantum supremacy [9]–[11].”
BackgroundA Simple Proof that Toffoli and Hadamard are Quantum Universal
Aharonov, D. (2003) · arXiv (Cornell University) · doi:10.48550/arxiv.quant-ph/0301040 · Cited by 123 · cited 1×
“The nearly universal set contains only two types of gates: the Toffoli Gate and the Hadamard Gate [28] [29].”
BackgroundQuantum synchronization on the IBM Q system
Koppenhöfer, M., Bruder, C., & Roulet, A. (2020) · Physical Review Research · doi:10.1103/physrevresearch.2.023026 · Cited by 81 · cited 1×
“, quantum oscillator synchronization [15], variational quantum algorithms [16], and quantum search [17], and optimization approaches to convert common quantum circuits descriptions into the IBM QX have been proposed [18].”
MethodFrom Principal Subspaces to Principal Components with Linear Autoencoders
Plaut, E. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1804.10253 · Cited by 80
The Complexity of Approximating complex-valued Ising and Tutte partition functions
Goldberg, L. A., & Guo, H. (2017) · computational complexity · doi:10.1007/s00037-017-0162-2 · Cited by 49 · cited 1×
“The exact evaluation of the inner-product between two states created from a similar circuit with only one diagonal layer U Φ( (cid:126)x ) is P -hard [26].”
BackgroundQuantum Supremacy Is Both Closer and Farther than It Appears
Markov, I. L., Fatima, A., Isakov, S. V., & Boixo, S. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1807.10749 · Cited by 46 · cited 1×
“in terms of learning efficiency, or even exponentially in performance [10].”
BackgroundIntrusion Detection System Using PCA and Kernel PCA Methods
Elkhadir, Z., Chougdali, K., & Benattou, M. (2016) · Lecture notes in electrical engineering · doi:10.1007/978-3-319-30298-0_50 · Cited by 32 · cited 1×
“Principal component analysis (PCA) is arguably the most used technique for dimensionality reduction, so it has already been used in the context of intrusion detection [19], [20].”
MethodA taxonomy of parallel techniques for intrusion detection
Wheeler, P. S., & Fulp, E. W. (2007) · ACM Southeast Regional Conference (ACMSE) · doi:10.1145/1233341.1233391 · Cited by 26 · cited 1×
“The detection of security violations using machine learning (ML) has been extensively investigated [1]–[8].”
BackgroundClassical verification of quantum circuits containing few basis changes
Demarie, T. F., Ouyang, Y., & Fitzsimons, J. F. (2018) · Physical Review A · doi:10.1103/physreva.97.042319 · Cited by 17 · cited 1×
“The experimental implementation of the parameterized diagonal two-Qbit operations is done using CNOTs and Z − gates [23]. defect, the simulation of a single layer preparation circuit can be effectively performed. by uniform sampling [27].”
MethodEchoing the recent Google success: Foundational Roots of Quantum SupremacyPreprint
Khrennikov, A. Y. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1911.10337 · Cited by 12
Ruta: Implementations of neural autoencoders in R
Charte, D., Herrera, F., & Charte, F. (2019) · Knowledge-Based Systems · doi:10.1016/j.knosys.2019.01.014 · Cited by 8 · cited 1×
“3) Encoding of the data: we considered an implementation based on the autoencode() function provided by the Ruta package [31], [32] for the R statistical computing package”
MethodConstant Time Quantum search Algorithm Over A Datasets: An Experimental Study Using IBM Q ExperiencePreprint
Das, K., & Sadhu, A. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1810.03390 · Cited by 2 · cited 1×
“, quantum oscillator synchronization [15], variational quantum algorithms [16], and quantum search [17], and optimization approaches to convert common quantum circuits descriptions into the IBM QX have been proposed [18].”
Method
Research works printed but not linked: 1
- 2019, R Core Team R A language and environment for statistical computing
Cited by
Ferrag, M. A., Shu, L., Friha, O., & Yang, X. K. (2021) · IEEE/CAA Journal of Automatica Sinica · doi:10.1109/jas.2021.1004344 · Cited by 195
Quantum deep learning-based anomaly detection for enhanced network security
Hdaib, M., Rajasegarar, S., & Pan, L. (2024) · Quantum Machine Intelligence · doi:10.1007/s42484-024-00163-2 · Cited by 163
“One of the works in Gouveia and Correia (2020) compared a quantum support vector machine (QSVM) and a traditional support vector machine (SVM) to compare its performance in classifying the attack classes using the NSL-KDD (Shiravi et al. 2012) dataset, an enhanced variant of the KDD99 dataset (MIT…”
Security intrusion detection using quantum machine learning techniques
Kalinin, M., & Krundyshev, V. (2022) · Journal of Computer Virology and Hacking Techniques · doi:10.1007/s11416-022-00435-0 · Cited by 118
“[18] presents the experimental study of the QSVM runningwith conventional NSL-KDD andNB-15 datasets, that denotes the QSVM’s readiness to the intrusion detection on common attack datasets.”
MethodRanga, D., Rana, A., Prajapat, S., Kumar, P., Kumar, K., & Vasilakos, A. V. (2024) · Mathematics · doi:10.3390/math12213318 · Cited by 149
Al-Hawawreh, M. S., & Hossain, M. S. (2023) · Information Fusion · doi:10.1016/j.inffus.2023.101889 · Cited by 102
Metadata
- Authors
- Gouveia, A., & Correia, M.
- Affiliations
- Instituto de Engenharia de Sistemas e Computadores Investigação e Desenvolvimento (PT)
- ORCID
- Correia, Miguel
- Venue
- 2020 IEEE 19th International Symposium on Network Computing and Applications (NCA)
- Publisher
- IEEE
- Published
- 2020-11-24
- Pages
- 1-8
- OpenAlex
- W3118870888
- Cited by
- 52
- References
- 34
- Citations by year
- 2026: 14 · 2025: 16 · 2024: 13 · 2023: 5 · 2022: 2 · 2021: 2
- FWCI
- 1.89
- Type
- conference paper
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE International Symposium on Network Computing and Applications (NCA) (publishedVersion)
- Funders
- European Commission (830892 (SPARTA))
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › Network Security and Intrusion Detection
- Physical Sciences › Computer Science › Artificial Intelligence › Quantum Computing Algorithms and Architecture
- Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
- Keywords
- network intrusion detection, quantum machine learning, unsupervised learning, quantum information, support vector machine, malicious activity detection
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.