pith. sign in

arxiv: 2605.21164 · v1 · pith:FXN5WNKVnew · submitted 2026-05-20 · 💻 cs.LG · quant-ph

Q-SYNTH: Hybrid Quantum-Classical Adversarial Augmentation for Imbalanced Fraud Detection

Adam Innan , Mansour El Alami , Nouhaila Innan , Muhammad Shafique , Mohamed Bennai This is my paper

Pith reviewed 2026-05-21 05:47 UTC · model grok-4.3

classification 💻 cs.LG quant-ph
keywords quantum GANfraud detectionimbalanced learningsynthetic data generationhybrid quantum classicaltabular datagenerative models
0
0 comments X

The pith

A hybrid quantum-classical GAN generates synthetic fraud samples with reduced distributional mismatch compared to classical baselines.

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

Credit card fraud detection suffers from severe imbalance where fraudulent cases are rare. Q-SYNTH addresses this by using a parameterized quantum circuit to generate additional fraudulent transaction samples in a GAN setup. The quantum generator is paired with a classical neural network discriminator to create synthetic data that better matches the real fraud distribution. Evaluations show lower mismatch in key statistical measures while downstream classifiers perform similarly to those trained with classical synthetic data. This demonstrates that hybrid quantum approaches can be applied to practical imbalanced tabular data problems.

Core claim

Q-SYNTH employs a parameterized quantum circuit as the generator in a generative adversarial network, with a classical neural network as the discriminator, to synthesize minority-class fraud samples from tabular data. Under the evaluation protocol, this hybrid model reduces marginal distribution mismatch relative to a classical GAN baseline according to Kolmogorov-Smirnov statistics and Wasserstein distances, while delivering competitive performance in fraud detection tasks across quantum and classical classifiers.

What carries the argument

Parameterized quantum circuit serving as the generator in a hybrid GAN framework for minority class synthesis.

Load-bearing premise

That improvements in distributional similarity metrics will lead to better operational fraud detection in real systems.

What would settle it

Deploying the generated samples in an actual credit card fraud monitoring system and measuring the change in fraud catch rate compared to classical augmentation methods.

Figures

Figures reproduced from arXiv: 2605.21164 by Adam Innan, Mansour El Alami, Mohamed Bennai, Muhammad Shafique, Nouhaila Innan.

Figure 1
Figure 1. Figure 1: Motivation summary comparing distribution similarity and classification performance for five augmentation methods: C-GAN [18], WGAN-GP [19], GAN, SMOTE, and Q-SYNTH. The Similarity score (blue) is a single value in [0,1] formed by averaging two normalized signals: an inverted KS statistic (lower KS is better) and the KS p-value (higher is better). The Utility score (orange) is the downstream fraud Recall R… view at source ↗
Figure 2
Figure 2. Figure 2: Overview of the proposed Q-SYNTH pipeline for fraud data generation and evaluation. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Quantum generator circuit. It starts with angle embedding, followed by trainable single-qubit rotations (RX, RY, RZ) [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Generator loss over training epochs. We examine the training dynamics to assess whether adver￾sarial optimization remains qualitatively stable and whether the generator progressively improves its approximation of the target fraud distribution in the learned representation space. The generator loss trend, presented in [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Discriminator loss over training epochs. [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: ROC curve of an external logistic regression detector [PITH_FULL_IMAGE:figures/full_fig_p011_6.png] view at source ↗
read the original abstract

Credit card fraud detection is fundamentally challenged by extreme class imbalance, where fraudulent transactions are rare yet operationally critical. This imbalance often biases supervised learners toward the legitimate class, leading to high overall accuracy but weaker fraud-class recall and F1-score. This paper introduces Q-SYNTH, a hybrid classical--quantum generative adversarial framework in which a parameterized quantum circuit serves as the generator and a classical neural network serves as the discriminator. Q-SYNTH is designed for minority-class fraud synthesis in tabular data and is evaluated along two dimensions: statistical fidelity to real fraud samples and downstream performance for fraud detection. To this end, generated samples are assessed using distributional similarity measures based on Kolmogorov-Smirnov statistics and Wasserstein distances, real-vs-synthetic detectability measured by AUC-ROC, and downstream classification performance across both quantum and classical classifiers. Under the reported protocol, Q-SYNTH reduces marginal distribution mismatch relative to a classical GAN baseline while maintaining competitive downstream fraud-detection performance. Although SMOTE achieves the strongest feature-wise similarity and the classical GAN attains the highest downstream performance in several settings, Q-SYNTH offers a favorable compromise between distributional fidelity and downstream performance, supporting the feasibility of hybrid quantum augmentation for imbalanced fraud detection.

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

2 major / 1 minor

Summary. The manuscript introduces Q-SYNTH, a hybrid quantum-classical generative adversarial framework for minority-class augmentation in tabular credit-card fraud detection. A parameterized quantum circuit serves as the generator and a classical neural network as the discriminator. Synthetic samples are evaluated on marginal distributional fidelity (Kolmogorov-Smirnov statistics and Wasserstein distances), real-vs-synthetic detectability (AUC-ROC), and downstream classification performance on both quantum and classical models. The central claim is that Q-SYNTH reduces marginal distribution mismatch relative to a classical GAN baseline while remaining competitive on downstream fraud-detection tasks, thereby offering a favorable compromise between fidelity and utility and supporting the feasibility of hybrid quantum augmentation.

Significance. If the central claims are substantiated with quantitative results and joint-distribution validation, the work would provide an early demonstration of quantum generators within adversarial augmentation pipelines for highly imbalanced tabular data. The dual evaluation axis (distributional metrics plus downstream classifiers) is a constructive design choice, and the explicit positioning against both classical GAN and SMOTE baselines allows direct comparison. The practical relevance to fraud detection adds weight, though the current marginal-only focus limits the strength of any operational claims.

major comments (2)
  1. [Abstract] Abstract: the abstract asserts that Q-SYNTH reduces marginal distribution mismatch relative to a classical GAN baseline and maintains competitive downstream performance, yet supplies no quantitative results, error bars, specific experimental details, dataset sizes, or data-exclusion rules. This leaves the headline claims with minimal verifiable support.
  2. [Evaluation protocol] Evaluation protocol (as described in the abstract and results): the reported improvements rest exclusively on marginal distributional metrics (Kolmogorov-Smirnov statistics and Wasserstein distances). No evidence is presented that joint feature dependencies, pairwise correlations, or conditional distributions among the minority-class samples are preserved. For tabular fraud data, downstream classifier utility depends on realistic feature interactions; distortion of these joints could render the competitive performance attributable to the classical discriminator or post-processing rather than genuine augmentation quality.
minor comments (1)
  1. [Abstract] Abstract: the statement that SMOTE achieves the strongest feature-wise similarity and the classical GAN attains the highest downstream performance in several settings would benefit from explicit identification of those settings and the magnitude of the differences.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback on our manuscript. We address each major comment point by point below, indicating planned revisions where appropriate.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the abstract asserts that Q-SYNTH reduces marginal distribution mismatch relative to a classical GAN baseline and maintains competitive downstream performance, yet supplies no quantitative results, error bars, specific experimental details, dataset sizes, or data-exclusion rules. This leaves the headline claims with minimal verifiable support.

    Authors: We agree that the abstract would be strengthened by the inclusion of key quantitative results. In the revised manuscript we will update the abstract to report the specific reductions in Kolmogorov-Smirnov statistics and Wasserstein distances relative to the classical GAN baseline, together with the downstream performance metrics (e.g., F1-score or AUC-ROC on the fraud class) and brief details on dataset size and any preprocessing or exclusion rules applied. This will make the headline claims directly verifiable from the abstract. revision: yes

  2. Referee: [Evaluation protocol] Evaluation protocol (as described in the abstract and results): the reported improvements rest exclusively on marginal distributional metrics (Kolmogorov-Smirnov statistics and Wasserstein distances). No evidence is presented that joint feature dependencies, pairwise correlations, or conditional distributions among the minority-class samples are preserved. For tabular fraud data, downstream classifier utility depends on realistic feature interactions; distortion of these joints could render the competitive performance attributable to the classical discriminator or post-processing rather than genuine augmentation quality.

    Authors: We acknowledge that marginal metrics alone do not fully characterize joint distributions, which are important for tabular data. Our evaluation protocol deliberately emphasized marginal fidelity because the parameterized quantum circuit is configured to model per-feature distributions; full joint modeling remains computationally demanding on current quantum hardware. The competitive downstream performance observed on both quantum and classical classifiers provides indirect support that the generated samples preserve sufficient structure for practical utility, as severe joint distortion would be expected to degrade classifier metrics. To directly address the concern, we will add pairwise correlation comparisons and selected conditional distribution checks between real and synthetic minority-class samples in the revised results section. revision: partial

Circularity Check

0 steps flagged

Empirical comparison with no self-referential derivations or fitted predictions

full rationale

The paper reports an experimental hybrid quantum-classical GAN framework evaluated on tabular fraud data via Kolmogorov-Smirnov statistics, Wasserstein distances, AUC-ROC detectability, and downstream classifier performance against classical GAN and SMOTE baselines. No equations, uniqueness theorems, ansatzes, or parameter-fitting steps are described that reduce a claimed prediction or result to the input data or self-citation by construction. The central feasibility claim rests on reported empirical outcomes rather than any load-bearing derivation chain, making the work self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

Review is based solely on the abstract; specific free parameters, axioms, and training details are not described. The approach implicitly relies on the expressivity of parameterized quantum circuits for tabular distributions and standard GAN training dynamics.

free parameters (1)
  • Parameterized quantum circuit weights
    Trainable parameters of the quantum generator that are optimized during adversarial training to match fraud distributions.
axioms (1)
  • domain assumption Parameterized quantum circuits can effectively model distributions of rare tabular fraud events
    Required for the quantum generator to produce useful synthetic samples as claimed.

pith-pipeline@v0.9.0 · 5764 in / 1374 out tokens · 69651 ms · 2026-05-21T05:47:23.531710+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

57 extracted references · 57 canonical work pages · 2 internal anchors

  1. [1]

    [Online]

    Jul 2025. [Online]. Available: https://www.biocatch.com/press-release/ banking-scams-up-65-globally-in-last-year

    work page 2025

  2. [2]

    [Online]

    2025. [Online]. Available: https://www.financialprofessionals.org/ training-resources/resources/survey-research-economic-data/details/ payments-fraud

    work page 2025

  3. [3]

    [Online]

    Nov 2025. [Online]. Available: https://www.bbc.com/articles/ c74j00lzdpvo

    work page 2025

  4. [4]

    Deep learning in financial fraud detection: Innovations, challenges, and applications,

    Y . Chen, C. Zhao, Y . Xu, C. Nie, and Y . Zhang, “Deep learning in financial fraud detection: Innovations, challenges, and applications,”Data Science and Management, 2025

    work page 2025

  5. [5]

    A systematic review of ai-enhanced techniques in credit card fraud detection,

    I. Y . Hafez, A. Y . Hafez, A. Saleh, A. A. Abd El-Mageed, and A. A. Abohany, “A systematic review of ai-enhanced techniques in credit card fraud detection,”Journal of Big Data, vol. 12, no. 1, p. 6, 2025

    work page 2025

  6. [6]

    Generative modeling for imbalanced credit card fraud transaction detection,

    M. Tayebi and S. El Kafhali, “Generative modeling for imbalanced credit card fraud transaction detection,”Journal of Cybersecurity and Privacy, vol. 5, no. 1, p. 9, 2025

    work page 2025

  7. [7]

    Credit card fraud detection using deep learning techniques and handling unbalanced class distributions with agss,

    C. S. Nama and K. S. Banu, “Credit card fraud detection using deep learning techniques and handling unbalanced class distributions with agss,”IEEE Access, vol. 14, pp. 1847–1864, 2025

    work page 2025

  8. [8]

    Credit card fraud detection with adasyn oversampling and shap-based interpretability: A comparative ensemble approach,

    A. Malhotra, B. S. Hada, A. Mishra, M. S. A. Bashaet al., “Credit card fraud detection with adasyn oversampling and shap-based interpretability: A comparative ensemble approach,” in2025 6th International Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV). IEEE, 2025, pp. 891–899

    work page 2025

  9. [9]

    Mitigating class imbalance in banking transactions: A graph-based gan solution for fraud detection,

    G. KK, A. BV , H. P. Reddy, N. R. Mayur, and B. Bhowmik, “Mitigating class imbalance in banking transactions: A graph-based gan solution for fraud detection,”Computational Economics, pp. 1–53, 2026

    work page 2026

  10. [10]

    Mixed quantum–classical method for fraud detection with quantum feature selection,

    M. Grossi, N. Ibrahim, V . Radescu, R. Loredo, K. V oigt, C. V on Altrock, and A. Rudnik, “Mixed quantum–classical method for fraud detection with quantum feature selection,”IEEE Transactions on Quantum Engineering, vol. 3, pp. 1–12, 2022

    work page 2022

  11. [11]

    Financial fraud detection: a comparative study of quantum machine learning models,

    N. Innan, M. A.-Z. Khan, and M. Bennai, “Financial fraud detection: a comparative study of quantum machine learning models,”International Journal of Quantum Information, vol. 22, no. 02, p. 2350044, 2024

    work page 2024

  12. [12]

    Financial fraud detection using quantum graph neural networks,

    N. Innan, A. Sawaika, A. Dhor, S. Dutta, S. Thota, H. Gokal, N. Patel, M. A.-Z. Khan, I. Theodonis, and M. Bennai, “Financial fraud detection using quantum graph neural networks,”Quantum Machine Intelligence, vol. 6, no. 1, p. 7, 2024

    work page 2024

  13. [13]

    QFNN-FFD: Quantum federated neural network for financial fraud detection,

    N. Innan, A. Marchisio, M. Bennai, and M. Shafique, “QFNN-FFD: Quantum federated neural network for financial fraud detection,” in2025 IEEE International Conference on Quantum Software (QSW). IEEE, 2025, pp. 41–47

    work page 2025

  14. [14]

    Comparative performance analysis of quantum machine learning architectures for credit card fraud detection,

    M. El Alami, N. Innan, M. Shafique, and M. Bennai, “Comparative performance analysis of quantum machine learning architectures for credit card fraud detection,”Applied Intelligence, vol. 56, no. 3, p. 83, 2026

    work page 2026

  15. [15]

    A privacy-preserving federated framework with hybrid quantum-enhanced learning for financial fraud detection,

    A. Sawaika, S. Krishna, T. Tomar, D. P. Suggisetti, A. Lal, T. Shrivastav, N. Innan, and M. Shafique, “A privacy-preserving federated framework with hybrid quantum-enhanced learning for financial fraud detection,” in2025 IEEE International Conference on Quantum Computing and Engineering (QCE), vol. 1. IEEE, 2025, pp. 1578–1588

    work page 2025

  16. [16]

    Circuithunt: Automated quantum circuit screening for superior credit-card fraud detection,

    N. Innan, A. Singh, and M. Shafique, “Circuithunt: Automated quantum circuit screening for superior credit-card fraud detection,” in2025 IEEE International Conference on Quantum Artificial Intelligence (QAI). IEEE, 2025, pp. 432–438

    work page 2025

  17. [17]

    Fid-qae: A fidelity-driven quantum autoencoder for credit card fraud detection,

    M. E. Alami, A. Innan, N. Innan, M. Shafique, and M. Bennai, “Fid-qae: A fidelity-driven quantum autoencoder for credit card fraud detection,” arXiv preprint arXiv:2512.12689, 2025

    work page arXiv 2025

  18. [18]

    Generative adversarial networks for transaction anomaly detection: A synthetic data approach,

    M. Hasan, K. N. Hasan, F. S. Tamim, M. S. Arefin, and A. W. Reza, “Generative adversarial networks for transaction anomaly detection: A synthetic data approach,” in2025 International Conference on Quantum Photonics, Artificial Intelligence, and Networking (QPAIN). IEEE, 2025, pp. 1–6

    work page 2025

  19. [19]

    Improved training of wasserstein gans,

    I. Gulrajani, F. Ahmed, M. Arjovsky, V . Dumoulin, and A. C. Courville, “Improved training of wasserstein gans,”Advances in neural information processing systems, vol. 30, 2017

    work page 2017

  20. [20]

    Classifier ensemble based on multiview optimization for high-dimensional imbalanced data classification,

    Y . Xu, Z. Yu, and C. P. Chen, “Classifier ensemble based on multiview optimization for high-dimensional imbalanced data classification,”IEEE Transactions on Neural Networks and Learning Systems, vol. 35, no. 1, pp. 870–883, 2022

    work page 2022

  21. [21]

    Learning a distance metric by balancing kl-divergence for imbalanced datasets,

    L. Feng, H. Wang, B. Jin, H. Li, M. Xue, and L. Wang, “Learning a distance metric by balancing kl-divergence for imbalanced datasets,” IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 49, no. 12, pp. 2384–2395, 2018

    work page 2018

  22. [22]

    Credit card fraud detection using state-of-the-art machine learning and deep learning algorithms,

    F. K. Alarfaj, I. Malik, H. U. Khan, N. Almusallam, M. Ramzan, and M. Ahmed, “Credit card fraud detection using state-of-the-art machine learning and deep learning algorithms,”Ieee Access, vol. 10, pp. 39 700– 39 715, 2022

    work page 2022

  23. [23]

    Deep learning for credit card fraud detection: A review of algorithms, challenges, and solutions,

    I. D. Mienye and N. Jere, “Deep learning for credit card fraud detection: A review of algorithms, challenges, and solutions,”IEEE Access, 2024

    work page 2024

  24. [24]

    Recent advances in credit card fraud detection: An analytical review of frameworks, methodologies, datasets, and challenges,

    T. A. Gaav, H. U. Adoga, and T. Moses, “Recent advances in credit card fraud detection: An analytical review of frameworks, methodologies, datasets, and challenges,”Journal of Future Artificial Intelligence and Technologies, vol. 2, no. 3, pp. 343–369, 2025

    work page 2025

  25. [25]

    Smote: synthetic minority over-sampling technique,

    N. V . Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, “Smote: synthetic minority over-sampling technique,”Journal of artificial intelligence research, vol. 16, pp. 321–357, 2002

    work page 2002

  26. [26]

    Borderline-smote: a new over- sampling method in imbalanced data sets learning,

    H. Han, W.-Y . Wang, and B.-H. Mao, “Borderline-smote: a new over- sampling method in imbalanced data sets learning,” inInternational conference on intelligent computing. Springer, 2005, pp. 878–887

    work page 2005

  27. [27]

    Adasyn: Adaptive synthetic sam- pling approach for imbalanced learning,

    H. He, Y . Bai, E. A. Garcia, and S. Li, “Adasyn: Adaptive synthetic sam- pling approach for imbalanced learning,” in2008 IEEE international joint conference on neural networks (IEEE world congress on computational intelligence). Ieee, 2008, pp. 1322–1328

    work page 2008

  28. [28]

    Time-aware density-based generative network for imbalanced data in credit card fraud detection,

    Y . Hong, Y . Xie, Y . Zheng, W. Li, T. Zhao, and X. Qiu, “Time-aware density-based generative network for imbalanced data in credit card fraud detection,” in2024 International Conference on Networking, Sensing and Control (ICNSC). IEEE, 2024, pp. 1–7

    work page 2024

  29. [29]

    Generative adversarial nets,

    I. J. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y . Bengio, “Generative adversarial nets,” Advances in neural information processing systems, vol. 27, 2014

    work page 2014

  30. [30]

    Using generative adversarial networks for improving classification effectiveness in credit card fraud detection,

    U. Fiore, A. De Santis, F. Perla, P. Zanetti, and F. Palmieri, “Using generative adversarial networks for improving classification effectiveness in credit card fraud detection,”Information Sciences, vol. 479, pp. 448– 455, 2019

    work page 2019

  31. [31]

    An efficient domain-adaptation method using gan for fraud detection,

    J. Hwang and K. Kim, “An efficient domain-adaptation method using gan for fraud detection,” 2020

    work page 2020

  32. [32]

    Comparative analysis of machine learning algorithms using gans through credit card fraud detection,

    E. Strelcenia and S. Prakoonwit, “Comparative analysis of machine learning algorithms using gans through credit card fraud detection,” in 2022 International Conference on Computing, Networking, Telecommuni- cations & Engineering Sciences Applications (CoNTESA). IEEE, 2022, pp. 1–5

    work page 2022

  33. [33]

    Ida-gan: A novel imbalanced data augmentation gan,

    H. Yang and Y . Zhou, “Ida-gan: A novel imbalanced data augmentation gan,” in2020 25th international conference on pattern recognition (ICPR). IEEE, 2021, pp. 8299–8305

    work page 2021

  34. [34]

    Smotified-gan for class imbalanced pattern classification problems,

    A. Sharma, P. K. Singh, and R. Chandra, “Smotified-gan for class imbalanced pattern classification problems,”Ieee Access, vol. 10, pp. 30 655–30 665, 2022

    work page 2022

  35. [35]

    Improving credit card fraud detection using machine learning and gan technology,

    N. T. Ali, S. J. Hasan, A. Ghandour, and Z. S. Al-Hchimy, “Improving credit card fraud detection using machine learning and gan technology,” inBIO Web of Conferences, vol. 97. EDP Sciences, 2024, p. 00076

    work page 2024

  36. [36]

    Adversarial attack detection using explainable ai and generative models in real-time financial fraud monitoring systems,

    U. U. James, C. N. Idika, L. A. Enyejo, K. Abiodun, and J. O. Enyejo, “Adversarial attack detection using explainable ai and generative models in real-time financial fraud monitoring systems,”International Journal of Scientific Research and Modern Technology, vol. 3, no. 12, pp. 142–157, 2024

    work page 2024

  37. [37]

    Utilizing gans for fraud detection: model training with synthetic transaction data,

    M. Zhu, Y . Gong, Y . Xiang, H. Yu, and S. Huo, “Utilizing gans for fraud detection: model training with synthetic transaction data,” inInternational Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024), vol. 13180. SPIE, 2024, pp. 887–894

    work page 2024

  38. [38]

    A new gan-based data augmentation method for handling class imbalance in credit card fraud detection,

    E. Strelcenia and S. Prakoonwit, “A new gan-based data augmentation method for handling class imbalance in credit card fraud detection,” in 2023 10th International Conference on Signal Processing and Integrated Networks (SPIN). IEEE, 2023, pp. 627–634

    work page 2023

  39. [39]

    Fraud detection in banking using generative ai,

    S. N. P. Kumar, “Fraud detection in banking using generative ai,”Journal Of Engineering And Computer Sciences, vol. 4, no. 11, pp. 133–145, 2025

    work page 2025

  40. [40]

    Detection of ai deepfake and fraud in online payments using gan-based models,

    Z. Ke, S. Zhou, Y . Zhou, C. H. Chang, and R. Zhang, “Detection of ai deepfake and fraud in online payments using gan-based models,” in 2025 8th International Conference on Advanced Algorithms and Control Engineering (ICAACE). IEEE, 2025, pp. 1786–1790

    work page 2025

  41. [41]

    A robust hybrid model integrating gans, xgboost, and reinforcement learning (rl),

    S. A. Mohsen, N. M. Munassar, and M. F. Abdullah, “A robust hybrid model integrating gans, xgboost, and reinforcement learning (rl),”Journal of Science and Technology, vol. 30, no. 9, 2025

    work page 2025

  42. [42]

    Gan_bert: An advanced neural architecture for effective fraud detection on imbalanced datasets,

    H. Wang, Y . Gong, and C. Yu, “Gan_bert: An advanced neural architecture for effective fraud detection on imbalanced datasets,” inProceedings of the 2025 6th International Conference on Computer Information and Big Data Applications, 2025, pp. 1434–1443

    work page 2025

  43. [43]

    Integrated feature-temporal gan for imbalanced transaction fraud detection,

    Y . Zheng, Y . Xie, and J. Yao, “Integrated feature-temporal gan for imbalanced transaction fraud detection,”IEEE Access, 2025

    work page 2025

  44. [44]

    Ensemble synthesized minority oversampling-based generative adversar- ial networks and random forest algorithm for credit card fraud detection,

    F. A. Ghaleb, F. Saeed, M. Al-Sarem, S. N. Qasem, and T. Al-Hadhrami, “Ensemble synthesized minority oversampling-based generative adversar- ial networks and random forest algorithm for credit card fraud detection,” IEEE Access, vol. 11, pp. 89 694–89 710, 2023

    work page 2023

  45. [45]

    Quantum machine learning,

    J. Biamonte, P. Wittek, N. Pancotti, P. Rebentrost, N. Wiebe, and S. Lloyd, “Quantum machine learning,”Nature, vol. 549, no. 7671, pp. 195–202, 2017

    work page 2017

  46. [46]

    An introduction to quantum machine learning,

    M. Schuld, I. Sinayskiy, and F. Petruccione, “An introduction to quantum machine learning,”Contemporary Physics, vol. 56, no. 2, pp. 172–185, 2015

    work page 2015

  47. [47]

    A survey of quantum machine learning: Foundations, algo- rithms, frameworks, data and applications,

    F. Rodríguez-Díaz, D. Gutiérrez-Avilés, A. Troncoso, and F. Martínez- Álvarez, “A survey of quantum machine learning: Foundations, algo- rithms, frameworks, data and applications,”ACM Computing Surveys, vol. 58, no. 4, pp. 1–35, 2025

    work page 2025

  48. [48]

    Quantum generative adversarial learning,

    S. Lloyd and C. Weedbrook, “Quantum generative adversarial learning,” Physical review letters, vol. 121, no. 4, p. 040502, 2018

    work page 2018

  49. [49]

    Quantum generative adversarial networks

    P.-L. Dallaire-Demers and N. Killoran, “Quantum generative adversarial networks,”arXiv preprint arXiv:1804.08641, 2018

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  50. [50]

    Quantum generative adversarial networks for learning and loading random distributions,

    C. Zoufal, A. Lucchi, and S. Woerner, “Quantum generative adversarial networks for learning and loading random distributions,”npj Quantum Information, vol. 5, no. 1, p. 103, 2019

    work page 2019

  51. [51]

    Quantum generative modeling for financial time series with temporal correlations,

    D. Dechant, E. J. E. Schwander, L. Van Drooge, C. Moussa, D. Gar- laschelli, V . Dunjko, and J. Tura Brugués, “Quantum generative modeling for financial time series with temporal correlations,”Machine Learning: Science and Technology, 2026

    work page 2026

  52. [52]

    ihqgan: A lightweight invertible hybrid quantum-classical gener- ative adversarial networks for unsupervised image-to-image translation,

    X. Yang, R. Zhou, S. Jia, Y . Li, J. Yan, Z. Long, W. Guo, F. Xiong, and W. Xu, “ihqgan: A lightweight invertible hybrid quantum-classical gener- ative adversarial networks for unsupervised image-to-image translation,” Expert Systems with Applications, vol. 296, p. 128865, 2026

    work page 2026

  53. [53]

    Survey of quantum generative adversarial networks (qgan) to generate images,

    M. Pajuhanfard, R. Kiani, and V . S. Sheng, “Survey of quantum generative adversarial networks (qgan) to generate images,”Mathematics, vol. 12, no. 23, p. 3852, 2024

    work page 2024

  54. [54]

    Image denoising using quantum deep convolutional generative adversarial network for medical images,

    P. Nandal, S. Pahal, and G. M. Upadhyay, “Image denoising using quantum deep convolutional generative adversarial network for medical images,”International Journal of Computational Intelligence Systems, vol. 18, no. 1, p. 190, 2025

    work page 2025

  55. [55]

    Data augmentation in cancer image classification problem with quantum gan,

    M. B. Andra, A. R. Yuliani, J. A. Kadar, A. Ramdan, and H. F. Pardede, “Data augmentation in cancer image classification problem with quantum gan,”Quantum Machine Intelligence, vol. 7, no. 2, p. 90, 2025

    work page 2025

  56. [56]

    Credit card fraud detection,

    M. L. G. ULB, “Credit card fraud detection,” Mar 2018. [Online]. Available: https://www.kaggle.com/datasets/mlg-ulb/creditcardfraud

    work page 2018

  57. [57]

    PennyLane: Automatic differentiation of hybrid quantum-classical computations

    V . Bergholm, J. Izaac, M. Schuld, C. Gogolin, S. Ahmed, V . Ajith, M. S. Alam, G. Alonso-Linaje, B. AkashNarayanan, A. Asadiet al., “Pennylane: Automatic differentiation of hybrid quantum-classical computations,” arXiv preprint arXiv:1811.04968, 2018

    work page internal anchor Pith review Pith/arXiv arXiv 2018