pith. sign in

arxiv: 2605.22527 · v1 · pith:C45SJZEKnew · submitted 2026-05-21 · 💻 cs.NE · cs.ET

Quantum Genetic Optimization for Negative Selection Algorithms in Anomaly Detection

Giancarlo P. Gamberi , Calebe P. Bianchini This is my paper

Pith reviewed 2026-05-22 01:33 UTC · model grok-4.3

classification 💻 cs.NE cs.ET
keywords quantum genetic algorithmnegative selection algorithmanomaly detectionartificial immune systemdetector generationfinancial transactions dataset
0
0 comments X

The pith

Quantum genetic optimization improves anomaly detection accuracy

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper proposes the Quantum Genetic Negative Selection Algorithm (QGNSA) by integrating a quantum genetic algorithm into the existing EvoSeedRNSA method for generating detectors in anomaly detection. It leverages quantum superposition to explore multiple possibilities simultaneously and probabilistic amplitude adjustments to improve convergence in identifying self and non-self patterns. Evaluations on the Metaverse Financial Transactions Dataset show that this approach yields higher anomaly detection accuracy than the classical version and remains effective across different hyperparameter settings. A reader might care if quantum enhancements can make immune-inspired systems more practical for real-world security tasks like monitoring digital transactions.

Core claim

The paper establishes that replacing the classical evolutionary optimization in negative selection algorithms with a quantum genetic algorithm, which uses superposition and amplitude adjustment for detector generation, results in superior anomaly detection performance on the Metaverse Financial Transactions Dataset while maintaining robustness to hyperparameter variations.

What carries the argument

Quantum Genetic Algorithm (QGA) integrated into the detector generation process of EvoSeedRNSA, using quantum superposition and probabilistic amplitude adjustment to enhance search space exploration and convergence.

If this is right

  • QGNSA demonstrates superior anomaly detection accuracy compared to classical methods on financial transaction data.
  • The quantum approach maintains robustness under varying hyperparameter configurations.
  • Quantum computing techniques can provide advantages in artificial immune systems for high-dimensional anomaly detection tasks.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the quantum advantage persists, it could extend to other evolutionary algorithms in machine learning for improved exploration.
  • Deploying on actual quantum hardware might reveal practical efficiency gains beyond simulations.
  • Hybrid quantum-classical methods could balance computational cost with performance benefits in anomaly detection systems.

Load-bearing premise

That the quantum features of superposition and amplitude adjustment actually improve the search for effective detectors beyond what classical genetic algorithms can achieve.

What would settle it

A direct comparison experiment on the Metaverse Financial Transactions Dataset where the quantum version shows no improvement in accuracy or robustness over the classical EvoSeedRNSA would falsify the main claim.

Figures

Figures reproduced from arXiv: 2605.22527 by Calebe P. Bianchini, Giancarlo P. Gamberi.

Figure 1
Figure 1. Figure 1: Diagram of the composition of a population ( [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Confusion matrix for the first Quantum Algorithm execution. [PITH_FULL_IMAGE:figures/full_fig_p011_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Confusion matrix for the Classical algorithm execution. [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: This indicates that it is highly effective at detecting anomalies, successfully identifying [PITH_FULL_IMAGE:figures/full_fig_p013_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Confusion matrix of the second quantum test. [PITH_FULL_IMAGE:figures/full_fig_p014_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Confusion matrix of the third quantum test. [PITH_FULL_IMAGE:figures/full_fig_p014_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Confusion matrix of the fourth quantum test. [PITH_FULL_IMAGE:figures/full_fig_p014_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Confusion matrix of the second classical test. [PITH_FULL_IMAGE:figures/full_fig_p015_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Confusion matrix of the third classical test. [PITH_FULL_IMAGE:figures/full_fig_p015_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Confusion matrix of the fourth classical test. [PITH_FULL_IMAGE:figures/full_fig_p015_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Metrics of the four tests of the quantum algorithm. [PITH_FULL_IMAGE:figures/full_fig_p016_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Metrics of the four tests of the classical algorithm. [PITH_FULL_IMAGE:figures/full_fig_p016_12.png] view at source ↗
read the original abstract

Negative Selection Algorithms (NSAs), inspired by the self/non-self discrimination mechanism of the human immune system, have been widely employed in anomaly detection. However, their effectiveness is often constrained by the efficiency of detector generation. This paper presents the Quantum Genetic Negative Selection Algorithm (QGNSA), a novel approach that integrates a Quantum Genetic Algorithm (QGA) into the EvoSeedRNSA algorithm, replacing its classical evolutionary optimization process. The proposed method exploits quantum superposition and probabilistic amplitude adjustment to enhance search space exploration and convergence efficiency in the detector generation process. Empirical evaluations using the Metaverse Financial Transactions Dataset demonstrate that QGNSA achieves superior anomaly detection accuracy compared to its classical counterpart while maintaining robustness under varying hyperparameter configurations. The experimental results highlight the potential advantages of quantum computing in artificial immune systems, particularly in high-dimensional anomaly detection tasks. Future research will focus on further optimizing quantum circuit design, deploying the algorithm on real quantum hardware, and exploring hybrid quantum-classical approaches for improved computational efficiency.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 1 minor

Summary. The manuscript proposes the Quantum Genetic Negative Selection Algorithm (QGNSA) by integrating a Quantum Genetic Algorithm into the EvoSeedRNSA framework for anomaly detection in high-dimensional data. It replaces classical evolutionary optimization with quantum superposition and probabilistic amplitude adjustment to improve detector generation efficiency. The central claim is that empirical evaluations on the Metaverse Financial Transactions Dataset show QGNSA achieves superior anomaly detection accuracy and robustness to hyperparameter changes relative to its classical counterpart.

Significance. If the claimed performance gains are rigorously demonstrated, this could indicate practical value for quantum-inspired methods in artificial immune systems, particularly for search-space exploration in anomaly detection tasks like financial transaction monitoring. The work would then contribute by linking quantum genetic optimization to negative selection algorithms, potentially motivating hybrid quantum-classical implementations.

major comments (1)
  1. [Abstract] Abstract: The claim that 'Empirical evaluations using the Metaverse Financial Transactions Dataset demonstrate that QGNSA achieves superior anomaly detection accuracy compared to its classical counterpart while maintaining robustness under varying hyperparameter configurations' is unsupported by any quantitative evidence. No accuracy, F1, AUC, or other metrics are reported, no direct comparisons to EvoSeedRNSA or other baselines are provided, and no details on experimental setup, statistical tests, or the specific contribution of quantum superposition/probabilistic amplitude adjustment appear. This is load-bearing for the paper's central empirical claim.
minor comments (1)
  1. [Abstract] Abstract: The phrasing 'the experimental results highlight the potential advantages' is vague without any results shown; consider rephrasing to reflect that results are forthcoming or to be detailed in the full manuscript.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their thorough review and valuable feedback on our manuscript. We address the major comment point by point below and will incorporate revisions where appropriate to strengthen the paper.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The claim that 'Empirical evaluations using the Metaverse Financial Transactions Dataset demonstrate that QGNSA achieves superior anomaly detection accuracy compared to its classical counterpart while maintaining robustness under varying hyperparameter configurations' is unsupported by any quantitative evidence. No accuracy, F1, AUC, or other metrics are reported, no direct comparisons to EvoSeedRNSA or other baselines are provided, and no details on experimental setup, statistical tests, or the specific contribution of quantum superposition/probabilistic amplitude adjustment appear. This is load-bearing for the paper's central empirical claim.

    Authors: We agree that the abstract does not currently provide quantitative metrics or details to substantiate the performance claims. This is an important point for clarity. In the revised version of the manuscript, we will update the abstract to include specific quantitative evidence, such as reported accuracy improvements (e.g., percentage gains), F1-scores, AUC values, and direct comparisons to the classical EvoSeedRNSA and other baselines. We will also add a concise description of the experimental setup, mention the use of statistical tests for robustness, and highlight the contributions of quantum superposition and probabilistic amplitude adjustment. The full experimental results and analysis are detailed in the body of the paper, but we will ensure the abstract is self-supporting. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical claim with no derivations or self-referential definitions

full rationale

The provided abstract contains no equations, derivations, or mathematical steps. The central claim of superior anomaly detection accuracy is presented strictly as an empirical outcome from evaluations on the Metaverse Financial Transactions Dataset, not as a quantity defined in terms of fitted parameters, self-citations, or ansatzes. No load-bearing steps reduce by construction to inputs, and the text does not invoke uniqueness theorems or rename known results. The derivation chain is therefore self-contained as a standard empirical assertion.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

Only the abstract is available, so the ledger reflects claims stated there. The central claim rests on the unverified assumption that quantum mechanisms provide measurable search advantages in this specific NSA application, with no independent evidence or parameter details supplied.

free parameters (1)
  • hyperparameters
    The abstract states robustness under varying hyperparameter configurations but supplies no specific names or values.
axioms (1)
  • domain assumption Quantum superposition and probabilistic amplitude adjustment enhance search space exploration and convergence in detector generation
    Invoked in the abstract as the mechanism behind the claimed efficiency gains.

pith-pipeline@v0.9.0 · 5672 in / 1252 out tokens · 101879 ms · 2026-05-22T01:33:35.920450+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

20 extracted references · 20 canonical work pages

  1. [1]

    Hegazy, Mahmoud Abdel-Aty, and Salah S.A

    Mohammed Zidan, Salem F. Hegazy, Mahmoud Abdel-Aty, and Salah S.A. Obayya. Rapid solution of logical equivalence problems by quantum computation algorithm.Applied Soft Computing, 132:109844, 2023

    work page 2023

  2. [2]

    Evolutionary algorithms and quantum computing: Recent advances, opportunities, and challenges.IEEE Access, 13:16649–16670, 2025

    Junaid Ur Rehman, Muhammad Shohibul Ulum, Abdurrahman Wachid Shaffar, Amirul Adlil Hakim, Mujirin, Zaid Abdullah, Hayder Al-Hraishawi, Symeon Chatzinotas, and Hyundong Shin. Evolutionary algorithms and quantum computing: Recent advances, opportunities, and challenges.IEEE Access, 13:16649–16670, 2025

    work page 2025

  3. [3]

    Quantum-inspired evolutionary programming for economic facts allocation in power systems: Advancing quantum computing applications

    Arman Riaz Ochi, Syed Golam Mahmud, Bashudeb Chandra Ghosh, and Martin Mar- gala. Quantum-inspired evolutionary programming for economic facts allocation in power systems: Advancing quantum computing applications. In2024 IEEE 24th International Conference on Nanotechnology (NANO), pages 375–380, 2024

    work page 2024

  4. [4]

    Quantum machine learning algorithms for anomaly detection: A review.Future Generation Computer Systems, 166:107632, 2025

    SebastianoCorli, LorenzoMoro, DanieleDragoni, MassimilianoDispenza, andEnricoPrati. Quantum machine learning algorithms for anomaly detection: A review.Future Generation Computer Systems, 166:107632, 2025

    work page 2025

  5. [5]

    Negative selection algorithm research and applications in the last decade: A review.IEEE Transactions on Artificial Intelligence, 3(2):110–128, April 2022

    Kishor Datta Gupta and Dipankar Dasgupta. Negative selection algorithm research and applications in the last decade: A review.IEEE Transactions on Artificial Intelligence, 3(2):110–128, April 2022. 17

    work page 2022

  6. [6]

    Cited by: 0

    17th international conference on hybrid intelligent systems, his 2017.Advances in Intelligent Systems and Computing, 734, 2018. Cited by: 0

    work page 2017

  7. [7]

    Advanced Materials Research, 871, 2014

    2013 international conference on applied mechanics, fluid and solid mechanics, amfsm 2013. Advanced Materials Research, 871, 2014. Cited by: 0

    work page 2013

  8. [8]

    Image enhancement research based on quantum immune leap frog.Chinese Journal of Liquid Crystals and Displays, 30(3):499 – 504, 2015

    Yuan Zhou and Ren-Zhi Zhang. Image enhancement research based on quantum immune leap frog.Chinese Journal of Liquid Crystals and Displays, 30(3):499 – 504, 2015. Cited by: 1

    work page 2015

  9. [9]

    Soliman and Amr Adly

    Omar S. Soliman and Amr Adly. Quantum-negative selection algorithm for associative classification. In2012 IEEE International Conference on Granular Computing, pages 418– 423, 2012

    work page 2012

  10. [10]

    Portfolio optimization in both long and shortsellingtradingusingtrendratiosandquantum-inspiredevolutionaryalgorithms.IEEE Access, 9:152115–152130, 2021

    Yao-Hsin Chou, Yu-Chi Jiang, and Shu-Yu Kuo. Portfolio optimization in both long and shortsellingtradingusingtrendratiosandquantum-inspiredevolutionaryalgorithms.IEEE Access, 9:152115–152130, 2021

    work page 2021

  11. [11]

    Fractional-order pid motion control for auv using cloud-model-based quantum genetic algorithm.IEEE Access, 7:124828–124843, 2019

    Junhe Wan, Bo He, Dianrui Wang, Tianhong Yan, and Yue Shen. Fractional-order pid motion control for auv using cloud-model-based quantum genetic algorithm.IEEE Access, 7:124828–124843, 2019

    work page 2019

  12. [12]

    Quantum-enhanced regression analysis using state-of-the-art qlsas and qipms

    MohammadhosseinMohammadisiahroudi, ZeguanWu, BrandonAugustino, TamásTerlaky, and Arielle Carr. Quantum-enhanced regression analysis using state-of-the-art qlsas and qipms. In2022 IEEE/ACM 7th Symposium on Edge Computing (SEC), pages 375–380, 2022

    work page 2022

  13. [13]

    Bearing fault diagnosis using quantum machine learning

    Misha Urooj Khan, Muhammad Ahmad Kamran, and Wajiha Rahim Khan. Bearing fault diagnosis using quantum machine learning. In2023 20th International Bhurban Conference on Applied Sciences and Technology (IBCAST), pages 1–6, 2023

    work page 2023

  14. [14]

    Paving the way to hybrid quantum–classical scientific workflows.Future Generation Computer Systems, 158:346–366, 2024

    Sandeep Suresh Cranganore, Vincenzo De Maio, Ivona Brandic, and Ewa Deelman. Paving the way to hybrid quantum–classical scientific workflows.Future Generation Computer Systems, 158:346–366, 2024

    work page 2024

  15. [15]

    Special collection on advances in quantum computing: Methods, algorithms, and systems.Future Generation Computer Systems, 163:107503, 2025

    Stefano Markidis, Michela Taufer, and Lucio Grandinetti. Special collection on advances in quantum computing: Methods, algorithms, and systems.Future Generation Computer Systems, 163:107503, 2025

    work page 2025

  16. [16]

    Genetic quantum algorithm and its application to combinatorial optimization problem

    Kuk-Hyun Han and Jong-Hwan Kim. Genetic quantum algorithm and its application to combinatorial optimization problem. InProceedings of the 2000 Congress on Evolutionary Computation. CEC00 (Cat. No.00TH8512), volume 2, pages 1354–1360 vol.2, 2000

    work page 2000

  17. [17]

    Generating an approximately optimal detector set by evolving random seeds

    Jie Zhang, Wenjian Luo, and Baoliang Xu. Generating an approximately optimal detector set by evolving random seeds. In2009 Eighth IEEE International Conference on Depend- able, Autonomic and Secure Computing, pages 162–168, 2009

    work page 2009

  18. [18]

    Evoseedrnsaii: An improved evolutionary algorithm for gen- erating detectors in the real-valued negative selection algorithms.Applied Soft Computing, 19:18–30, 2014

    Jie Zhang and Wenjian Luo. Evoseedrnsaii: An improved evolutionary algorithm for gen- erating detectors in the real-valued negative selection algorithms.Applied Soft Computing, 19:18–30, 2014

    work page 2014

  19. [19]

    On average time complexity of evolutionary negative selection algorithms for anomaly detection

    Baoliang Xu, Wenjian Luo, Xingxin Pei, Min Zhang, and Xufa Wang. On average time complexity of evolutionary negative selection algorithms for anomaly detection. InProceed- ings of the First ACM/SIGEVO Summit on Genetic and Evolutionary Computation, GEC ’09, page 631–638, New York, NY, USA, 2009. Association for Computing Machinery

    work page 2009

  20. [20]

    Metaverse financial transactions dataset

    Jack Ward. Metaverse financial transactions dataset. 18

    work page