Pith. sign in

REVIEW 5 major objections 5 minor 57 references

Noisy HQNNs: A Comprehensive Analysis of Noise Robustness in Hybrid Quantum Neural Networks

T0 review · 5 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Under five quantum noise channels and two image tasks, the QuanNN classifier survives better than the QCNN.

desk verdict Worth refereeing but not accepting as is: a plausible small-scale noise-robustness benchmark whose headline ranking rests on single unseeded runs and no defined robustness metric. read the letter →

arxiv 2505.03378 v1 pith:XFZWCZPV submitted 2025-05-06 quant-ph

Tasnim Ahmed , Alberto Marchisio , Muhammad Kashif , Muhammad Shafique This is my paper · ORCID
classification quant-ph MSC 81P6868T07
keywords hybridquantumneuralnetworksconvolutionalnetworkquanvolutionalnoiserobustnesschannelsNISQdevicesMNISTFashion-MNIST
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tries to establish that architecture choice, not just noise level, determines whether hybrid quantum-classical image classifiers survive noisy quantum hardware. To show this, it trains two four-qubit architectures, the Quanvolutional Neural Network and the Quantum Convolutional Neural Network, on four-class MNIST and Fashion-MNIST under five quantum noise channels at probabilities from 0.1 to 1.0. It reports that the QuanNN is more robust overall, staying accurate under phase flip and phase damping at all tested levels and under bit flip at probability 1.0, while the QCNN can outperform its noise-free version for bit flip, phase flip, and phase damping at high noise probabilities. If these findings hold, a device's noise profile should guide which HQNN architecture a practitioner chooses, and some noise can be tolerated or even exploited rather than merely mitigated.

What carries the argument

The load-bearing object is the noise-injected variational quantum circuit: a four-qubit parameterized circuit with rotation gates and entangling layers, with a quantum noise gate inserted after every parametric-gate block and entanglement block. The QuanNN uses three basic-entangling layers as a sliding quantum filter feeding classical post-processing, while the QCNN uses three strongly-entangling layers with classical pre-processing to downsize the input. Five noise channels, namely bit flip, phase flip, phase damping, amplitude damping, and depolarizing noise, are implemented as Kraus-operator channels at ten probabilities from 0.1 to 1.0, and models are trained and validated under the injected noise on a mixed-state simulator. This setup lets the paper attribute accuracy differences to noise type, intensity, and architecture rather than to hardware-specific error correlations.

What would settle it

Re-run every architecture--noise-channel--probability--dataset condition with at least ten random seeds under fixed hyperparameters and report accuracy distributions; if the QuanNN-over-QCNN ranking and the beneficial-noise windows, such as bit flip at $p=1.0$ and phase flip or phase damping at high probabilities, do not consistently recur, the central claim is refuted. A sharper check for the bit-flip case is to verify whether a 100% bit-flip channel, being a deterministic X applied after every gate block, merely reparameterizes the circuit into an equivalent noise-free circuit; if so, the observed robustness is a learned compensation rather than genuine noise tolerance.

Watch

Extended reading notes

Core claim

On the authors' own terms, the central discovery is a conditional robustness ranking: across all five noise channels, both datasets, and all noise probabilities studied, the QuanNN exhibits greater robustness than the QCNN. The QuanNN is insensitive to phase flip and phase damping noise at every intensity, and at 100% bit flip probability it returns to near-noise-free training accuracy, apparently because deterministic X gates become a learnable transformation. It degrades sharply under depolarizing and amplitude damping noise once the probability passes roughly 0.3 to 0.5. The QCNN, by contrast, benefits from noise in specific windows, namely bit flip at 1.0, phase flip at several high probabilities, and phase damping at 0.4, but otherwise degrades, and on Fashion-MNIST its accuracy fluctuates with no consistent trend. The paper reads these patterns as evidence that noise resilience is a joint property of circuit architecture, noise channel, and task complexity.

Load-bearing premise

The robustness ranking rests on the assumption that the single training run per noise condition is representative, so that accuracy differences reflect architecture and noise channel rather than random initialization, optimizer noise, or the separately tuned learning rates.

Editorial extensions

If this is right

  • A device whose dominant errors are phase flip or phase damping should favor the QuanNN architecture, since it stays near noise-free accuracy at all tested intensities.
  • A device with strong bit-flip errors at high probability is not necessarily unusable: both architectures adapt at $p=1.0$, with the QuanNN returning to near-baseline accuracy.
  • Depolarizing and amplitude damping noise are the main failure modes for both architectures, with collapse typically beginning between $0.3$ and $0.5$ probability, so mitigation effort should target these channels first.
  • Task complexity amplifies noise sensitivity: the same architecture and noise level that works on MNIST can fluctuate badly on Fashion-MNIST, so noise-robustness claims should be benchmarked on more than one task.
  • QCNN deployment should be noise-profile-aware, since the architecture is robust only at select noise levels and noise types, not across the board.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable extension of the bit-flip result: a 100% bit-flip channel applies a deterministic X to every qubit after every gate block, which is equivalent to a fixed unitary transformation of the circuit; checking whether the trained weights simply rotate to compensate would separate 'learning despite noise' from 'learning the noise.'
  • The phase-damping robustness of the QuanNN may be a measurement-basis artifact: if the final measurement is in the Z basis and phase damping only kills off-diagonal coherences, the measurement probabilities are unchanged; swapping in X-basis or Y-basis measurements would test whether robustness disappears.
  • The paper's single-run design suggests a natural safeguard: future comparisons should report seed-averaged accuracies, since the ranking claim depends on differences that may be smaller than optimization variance.
  • If the noise-benefit windows for the QCNN are reproducible, they could be exploited deliberately by injecting matched noise during training as a regularizer, rather than treating all noise as something to remove.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 5 minor

Summary. The paper presents an empirical comparison of two hybrid quantum-classical image classifiers, a Quanvolutional Neural Network (QuanNN) and a Quantum Convolutional Neural Network (QCNN), under five quantum noise channels (bit flip, phase flip, depolarization, phase damping, amplitude damping) with probabilities from 0.1 to 1.0. Experiments use four-class subsets of MNIST and Fashion-MNIST, with 500 training samples and 150 validation samples per condition, and report training and validation accuracy over 20 epochs. The central claim, stated in Section IV-D, is that the QuanNN exhibits greater robustness than the QCNN across all examined noise types and task demands, with the QuanNN robust to phase-related noise and to bit-flip noise at p=1.0, and the QCNN sometimes outperforming its noise-free baseline at specific noise levels. The paper concludes with guidelines for architecture selection under different device noise profiles.

Significance. If the claimed architecture-dependent robustness ranking were established, the paper would provide practically useful guidance for choosing between QuanNN and QCNN architectures on NISQ devices. The study has real strengths: it covers a broad matrix of five noise channels, ten probability values, two architectures, and two datasets; it extends prior 2-qubit noise studies to 4-qubit multi-class settings; and it compares against explicit noise-free baselines. However, the central comparative claim is not currently supported by the evidence as presented: every condition is a single unseeded training run, no robustness metric is defined, and several stated conclusions conflict with the reported curves. The paper is a reasonable starting point for an empirical study, but it requires additional statistical grounding and a tightening of the claims before it can be accepted as a comprehensive analysis.

major comments (5)
  1. [Section IV-D and Figures 4-7] The central claim that "QuanNN exhibits greater robustness than the QCNN across all examined noise types and task demands" is not supported because each reported condition is a single unseeded run. Table II lists no random seed or repetition count, and Section IV-A states only that learning rates were tuned separately on noise-free models. Given that the text itself describes QCNN curves as "highly irregular" and "with no discernible pattern" (Section IV-C-b and IV-C-d), a single trajectory can reflect initialization or optimizer luck rather than architecture-level robustness. The authors should rerun each condition with multiple seeds and report mean accuracy with error bars or interquartile ranges, and should provide per-condition summary tables rather than relying only on curves.
  2. [Section III-C / Section IV-D] "Noise robustness" is never operationalized. The paper does not define a metric such as mean accuracy across noise levels, accuracy drop relative to the noise-free baseline, area under the noise-accuracy curve, or a failure threshold. The ranking in Section IV-D is therefore a qualitative reading of visual curves. A pre-specified robustness measure is needed before the paper can claim that one architecture is more robust than another; otherwise the conclusion is not falsifiable from the reported data.
  3. [Abstract and Section IV-C-d] The abstract states that the QCNN "tends to benefit from noise injection by outperforming noise-free models for bit flip, phase flip, and phase damping at high noise probabilities," but the body reports the opposite for phase damping: Section IV-C-d states that the QCNN "fails to learn effectively" at phase damping levels 0.7-1.0 on MNIST and that Fashion-MNIST shows "no discernible pattern," with only p=0.4 on MNIST outperforming the noise-free model. The abstract overstates the findings and should be corrected to match the reported results.
  4. [Section II-B and Sections IV-B-a/IV-C-a] At p=1.0, the bit-flip and phase-flip channels become deterministic unitary operations (Pauli X and Pauli Z, respectively), not stochastic noise. Presenting these results as "robustness to noise" or "benefiting from noise" conflates adaptation to a fixed unitary gate with resilience to stochastic errors. The authors should either analyze the p<1 cases separately from the deterministic p=1 cases or explicitly relabel the p=1 results as deterministic gate operations rather than noise robustness.
  5. [Section III-A and Table II] The experimental setup uses only 500 training samples and 150 validation samples from four classes, with dataset-specific learning rates. With a balanced four-class subset, 25% accuracy is the chance level, yet the text describes curves flattening at around 25% without noting this is failure to learn. The small sample size also makes the "outperforms noise-free" observations (e.g., 90% validation accuracy under depolarization 0.1 on MNIST in Section IV-C-c) hard to interpret without confidence intervals. Please report balanced-class confirmation, confidence intervals for the accuracies, and the chance-level threshold.
minor comments (5)
  1. [Figure 4 and Figure 5 captions] The captions contain typos: "Trainnig Accuracy" should be "Training Accuracy", and "mot fluctuation" should be "more fluctuation".
  2. [Section II-B, Equations (4)-(5)] Phase damping and amplitude damping are parameterized by gamma in the Kraus operators, but the text refers to the same quantity as "probability p". Please clarify the relation between gamma and the probability values reported in the experiments.
  3. [Section III-B] The noise-injection description says noise gates are placed "after every parametric gate block and entanglement block," but it is not explicit whether a noise channel is applied after every individual gate or after each layer, and whether the same channel is applied to all qubits. Please specify the exact insertion rule.
  4. [Section IV-D] The bullet point stating that QuanNN shows robustness to bit flip "against non-deterministic noise" is contradicted by the fact that p=1.0 is a deterministic X operation; please rephrase to avoid the inconsistency.
  5. [Table II and Section IV-A] For reproducibility, please include the PennyLane version, the random seed(s), the number of repeated runs, and the exact hyperparameter search procedure used to select the two learning rates.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the noise-robustness findings are empirical comparisons against external benchmarks, and architecture choices inherited from prior work do not define the reported outcomes.

full rationale

The paper does not contain a derivation chain in which a predicted quantity reduces by construction to a fitted input. Its central claim that the QuanNN 'exhibits greater robustness than the QCNN across all examined noise types and task demands' is an empirical summary of measured training and validation accuracies on MNIST and Fashion-MNIST under externally defined noise channels. The use of the authors' prior work [42] to select 'best-performing configurations' supplies experimental hyperparameters and architecture choices, but it does not fix the noise-robustness rankings that were subsequently measured; those rankings come from running the models under injected noise. There is no self-definitional step, no fitted parameter renamed as a prediction, no uniqueness theorem imported from the authors to force a choice, and no known empirical result repackaged under new coordinates. Concerns about single unseeded runs, separately tuned learning rates, and the absence of error bars are legitimate reproducibility and statistical-support concerns, but they are not circularity: the comparison is still evaluated against external data and standard noise models rather than against the paper's own assumptions.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The paper adds no new physical entities or mathematical constructs. Its free parameters are dataset-specific learning rates; its conclusions rely on modeling assumptions about how injected noise represents real NISQ devices and how representative the selected small architectures are.

free parameters (2)
  • Learning rate (MNIST) = 0.01
    Chosen in preliminary noise-free experiments (Section IV-A) and then fixed for all noisy MNIST runs; this choice affects convergence speed and could favor some noise conditions.
  • Learning rate (Fashion-MNIST) = 0.005
    Chosen in preliminary noise-free experiments (Section IV-A) for Fashion-MNIST; similar implications as the MNIST learning rate.
assumptions (4)
  • domain assumption The Kraus operators in Eqs. 1-5 and PennyLane's default.mixed simulator faithfully represent the five physical noise channels.
    All results depend on these noise models; real device noise includes crosstalk, calibration drift, and non-Markovian effects not captured here.
  • domain assumption Placing a noise gate after every parametric gate and entanglement block, with a single uniform probability per channel, is a representative model of NISQ gate errors, decoherence, and readout errors.
    The paper states this placement 'simulates gate errors, decoherence, and readout errors' (Section III-B); on real hardware error rates differ per gate and qubit.
  • ad hoc to paper The QuanNN (3 basic entangling layers) and QCNN (3 strongly entangling layers) configurations are representative of their respective architecture families.
    The configurations were selected as the best-performing from the authors' prior work [42]; other layer counts or entangling patterns could rank differently under noise.
  • domain assumption Training on 500 samples from 4 classes for 20 epochs is sufficient to reveal noise robustness trends for these architectures.
    The study uses small subsets and a fixed epoch budget; conclusions about 'robustness across all noise types' may not generalize to larger data or longer training.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Noisy HQNNs: A Comprehensive Analysis of Noise Robustness in Hybrid Quantum Neural Networks." pith.science (2026). https://pith.science/paper/XFZWCZPV

@misc{pith2026250503378,
  author       = {Pith},
  title        = {Pith review of: Noisy HQNNs: A Comprehensive Analysis of Noise Robustness in Hybrid Quantum Neural Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XFZWCZPV}},
  note         = {Machine review of arXiv:2505.03378}
}
read the original abstract

Hybrid Quantum Neural Networks (HQNNs) offer promising potential of quantum computing while retaining the flexibility of classical deep learning. However, the limitations of Noisy Intermediate-Scale Quantum (NISQ) devices introduce significant challenges in achieving ideal performance due to noise interference, such as decoherence, gate errors, and readout errors. This paper presents an extensive comparative analysis of two HQNN algorithms, Quantum Convolutional Neural Network (QCNN) and Quanvolutional Neural Network (QuanNN), assessing their noise resilience across diverse image classification tasks. We systematically inject noise into variational quantum circuits using five quantum noise channels: Phase Flip, Bit Flip, Phase Damping, Amplitude Damping, and Depolarizing Noise. By varying noise probabilities from 0.1 to 1.0, we evaluate the correlation between noise robustness and model behavior across different noise levels. Our findings demonstrate that different noise types and levels significantly influence HQNN performance. The QuanNN shows robust performance across most noise channels for low noise levels (0.1 - 0.4), but succumbs to diverse effects of depolarizing and amplitude damping noise at probabilities between (0.5 - 1.0). However, the QuanNN exhibits robustness to bit flip noise at high probabilities (0.9 - 1.0). On the other hand, the QCNN tends to benefit from the noise injection by outperforming noise-free models for bit flip, phase flip, and phase damping at high noise probabilities. However, for other noise types, the QCNN shows gradual performance degradation as noise increases. These insights aim to guide future research in error mitigation strategies to enhance HQNN models in the NISQ era.

Figures

Figures reproduced from arXiv: 2505.03378 by the authors.

Figure 1
Figure 1. Performance variations of HQNNs under different noise types and task [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Circuit structure and data process pipeline of our chosen HQNN [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Our methodology. A comprehensive comparative analysis of two different HQNN models (QuanNN and QCNN) is performed with different noise gate [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Comparative performance analysis of the QuanNN on the MNIST dataset under various noise channels with 0.1 to 1.0 intensities, benchmarked against [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Comparative performance analysis of the QuanNN on the Fashion-MNIST dataset under various noise channels with 0.1 to 1.0 intensities, benchmarked [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Comparative performance analysis of the QCNN on the MNIST dataset under various noise channels with 0.1 to 1.0 intensities, benchmarked against the [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Comparative performance analysis of the QCNN on the Fashion-MNIST dataset under various noise channels with 0.1 to 1.0 intensities, benchmarked [PITH_FULL_IMAGE:figures/full_fig_p009_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

57 extracted references · 35 canonical work pages

  1. [48]

    Investigating the effect of noise on the training performance of hybrid quantum neural networks,

    M. Kashif, E. Sychiuco, and M. Shafique, “Investigating the effect of noise on the training performance of hybrid quantum neural networks,” in IJCNN, 2024

    work page 2024

  2. [1]

    Quantum supremacy using a programmable superconduct- ing processor,

    F. Arute et al., “Quantum supremacy using a programmable superconduct- ing processor,” Nature, 2019

    work page 2019

  3. [2]

    Evidence for the utility of quantum computing before fault tolerance,

    Y . Kim et al., “Evidence for the utility of quantum computing before fault tolerance,” Nature, 2023

    work page 2023

  4. [3]

    Quantum computational advantage using photons,

    H.-S. Zhong et al. , “Quantum computational advantage using photons,” Science, 2020

    work page 2020

  5. [4]

    Quantum machine learning in feature hilbert spaces,

    M. Schuld and N. Killoran, “Quantum machine learning in feature hilbert spaces,” Physical Review Letters , 2019

    work page 2019

  6. [5]

    Programming quantum neural networks on nisq systems: An overview of technologies and methodologies,

    S. Markidis, “Programming quantum neural networks on nisq systems: An overview of technologies and methodologies,” Entropy, 2023

    work page 2023

  7. [6]

    A leap among quantum computing and quantum neural networks: A survey,

    F. V . Massoli et al., “A leap among quantum computing and quantum neural networks: A survey,” CoRR, vol. abs/2107.03313, 2022

    arXiv 2022

  8. [7]

    Power of data in quantum machine learning,

    H. Huang et al. , “Power of data in quantum machine learning,” Nature Communications, vol. 12, p. 2631, may 2021

    work page 2021

Show all 57 references
  1. [8]

    An introduction to quantum machine learning,

    M. Schuld, I. Sinayskiy, and F. Petruccione, “An introduction to quantum machine learning,” Contemporary Physics, 2015

    2015

  2. [9]

    Quantum machine learning,

    J. Biamonte et al., “Quantum machine learning,” Nature, 2017

    2017

  3. [10]

    The impact of cost function globality and locality in hybrid quantum neural networks on nisq devices,

    M. Kashif and S. Al-Kuwari, “The impact of cost function globality and locality in hybrid quantum neural networks on nisq devices,” Machine Learning: Science and Technology , vol. 4, no. 1, p. 015004, 2023

    2023

  4. [11]

    Fedqnn: Federated learning using quantum neural net- works,

    N. Innan et al. , “Fedqnn: Federated learning using quantum neural net- works,” in IJCNN, 2024

    2024

  5. [12]

    Financial fraud detection using quantum graph neural networks,

    N. Innan et al. , “Financial fraud detection using quantum graph neural networks,” Quantum machine intelligence , 2024

    2024

  6. [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,” arXiv, 2024

    2024

  7. [14]

    Po-qa: A framework for portfolio optimization using quantum algorithms,

    K. Zaman et al. , “Po-qa: A framework for portfolio optimization using quantum algorithms,” in QCE, 2024

    2024

  8. [15]

    Quantum bayesian networks for machine learning in oil-spill detection,

    O. I. Siddiqui et al., “Quantum bayesian networks for machine learning in oil-spill detection,” arXiv preprint arXiv:2412.19843 , 2024

    2024

  9. [16]

    Lep-qnn: Loan eligibility prediction using quantum neural networks,

    N. Innan et al., “Lep-qnn: Loan eligibility prediction using quantum neural networks,” arXiv preprint arXiv:2412.03158 , 2024

    2024

  10. [17]

    Fl-qdsnns: Federated learning with quantum dynamic spiking neural networks,

    N. Innan, A. Marchisio, and M. Shafique, “Fl-qdsnns: Federated learning with quantum dynamic spiking neural networks,” arXiv, 2024

    2024

  11. [18]

    Quantum clustering for cybersecurity,

    W. El Maouaki et al. , “Quantum clustering for cybersecurity,” in QCE, 2024

    2024

  12. [19]

    Qadqn: Quantum attention deep q-network for financial market prediction,

    S. Dutta et al. , “Qadqn: Quantum attention deep q-network for financial market prediction,” in QCE, 2024

    2024

  13. [20]

    Review of some existing QML frameworks and novel hybrid classical-quantum neural networks realising binary classification for the noisy datasets,

    N. Schetakis et al., “Review of some existing QML frameworks and novel hybrid classical-quantum neural networks realising binary classification for the noisy datasets,” Sci. Rep., 2022

    2022

  14. [21]

    Quantum machine learning revolution in healthcare: A systematic review of emerging perspectives and applications,

    U. Ullah and B. Garcia-Zapirain, “Quantum machine learning revolution in healthcare: A systematic review of emerging perspectives and applications,” IEEE Access, 2024

    2024

  15. [22]

    Quantum computing in the NISQ era and beyond,

    J. Preskill, “Quantum computing in the NISQ era and beyond,” Quantum, 2018

    2018

  16. [23]

    Efficient separate quantification of state preparation errors and measurement errors on quantum computers and their mitigation,

    H. Yu and T.-C. Wei, “Efficient separate quantification of state preparation errors and measurement errors on quantum computers and their mitigation,” arXiv, 2023

    2023

  17. [24]

    Challenges and opportunities in quantum machine learning,

    M. Cerezo et al. , “Challenges and opportunities in quantum machine learning,” Nature Computational Science , 2022

    2022

  18. [25]

    A survey on quantum machine learning: Current trends, challenges, opportunities, and the road ahead,

    K. Zaman et al., “A survey on quantum machine learning: Current trends, challenges, opportunities, and the road ahead,” arXiv:2310.10315, 2023

    2023 arXiv

  19. [26]

    Pennylane: Automatic differentiation of hybrid quantum-classical compu- tations,

    “Pennylane: Automatic differentiation of hybrid quantum-classical compu- tations,” CoRR, vol. abs/1811.04968, 2022

    2022 arXiv

  20. [27]

    Cutting is all you need: Execution of large- scale quantum neural networks on limited-qubit devices,

    A. Marchisio et al. , “Cutting is all you need: Execution of large- scale quantum neural networks on limited-qubit devices,” arXiv preprint arXiv:2412.04844, 2024

    2024 arXiv

  21. [28]

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

    M. Mohammadisiahroudi et al. , “Quantum-enhanced regression analysis using state-of-the-art qlsas and qipms,” in SEC, 2022

    2022

  22. [29]

    Supervised learning with quantum-enhanced feature spaces,

    V . Havl ´ıˇcek et al. , “Supervised learning with quantum-enhanced feature spaces,” Nature, 2019

    2019

  23. [30]

    The inductive bias of quantum kernels,

    J. K ¨ubler et al. , “The inductive bias of quantum kernels,” Advances in Neural Information Processing Systems , 2021

    2021

  24. [31]

    Taking advantage of noise in quantum reservoir computing,

    L. Domingo, G. Carlo, and F. Borondo, “Taking advantage of noise in quantum reservoir computing,” Scientific Reports, 2023

    2023

  25. [32]

    Quantum boltzmann machine,

    M. H. Amin et al., “Quantum boltzmann machine,” Phys. Rev. X, 2018

    2018

  26. [33]

    Quantum generative adversarial networks,

    P.-L. Dallaire-Demers and N. Killoran, “Quantum generative adversarial networks,” Phys. Rev. A, 2018

    2018

  27. [34]

    Demonstrating quantum advantage in hybrid quantum neural networks for model capacity,

    M. Kashif and S. Al-Kuwari, “Demonstrating quantum advantage in hybrid quantum neural networks for model capacity,” in ICRC, 2022

    2022

  28. [35]

    The power of quantum neural networks,

    A. Abbas et al. , “The power of quantum neural networks,” Nature Com- putational Science, 2021

    2021

  29. [36]

    Hybrid classical-quantum machine learning based on dissipative two-qubit channels,

    E. Ghasemian and M. K. Tavassoly, “Hybrid classical-quantum machine learning based on dissipative two-qubit channels,” Sci. Rep., 2022

    2022

  30. [37]

    Classification with quantum neural networks on near term processors,

    E. Farhi and H. Neven, “Classification with quantum neural networks on near term processors,” arXiv preprint arXiv:1802.06002 , 2018

    2018 arXiv

  31. [38]

    Computational advantage in hybrid quantum neural networks: Myth or reality?,

    M. Kashif, A. Marchisio, and M. Shafique, “Computational advantage in hybrid quantum neural networks: Myth or reality?,” in DAC, 2025

    2025

  32. [39]

    Nrqnn: The role of observable selection in noise-resilient quantum neural networks,

    M. Kashif and M. Shafique, “Nrqnn: The role of observable selection in noise-resilient quantum neural networks,” in Grid, Cloud, and Cluster Com- puting; Quantum Technologies; and Modeling, Simulation and Visualization Methods, Springer Nature Switzerland, 2025

    2025

  33. [40]

    Designing robust quantum neural networks: Ex- ploring expressibility, entanglement, and control rotation gate selection for enhanced quantum models,

    W. El Maouaki et al. , “Designing robust quantum neural networks: Ex- ploring expressibility, entanglement, and control rotation gate selection for enhanced quantum models,” arXiv preprint arXiv:2411.11870 , 2024

    2024 arXiv

  34. [41]

    Design space exploration of hybrid quantum– classical neural networks,

    M. Kashif and S. Al-Kuwari, “Design space exploration of hybrid quantum– classical neural networks,” Electronics, vol. 10, no. 23, p. 2980, 2021

    2021

  35. [42]

    A comparative analysis of hybrid-quantum classical neural networks,

    K. Zaman et al. , “A comparative analysis of hybrid-quantum classical neural networks,” in World Congress in Computer Science, Computer Engineering & Applied Computing , 2024

    2024

  36. [43]

    Quantum convolutional neural network for classical data classification,

    T. Hur, L. Kim, and D. K. Park, “Quantum convolutional neural network for classical data classification,” Quantum Machine Intelligence , 2022

    2022

  37. [44]

    Studying the impact of quantum-specific hyperparameters on hybrid quantum-classical neural networks,

    K. Zaman et al., “Studying the impact of quantum-specific hyperparameters on hybrid quantum-classical neural networks,” in World Congress in Computer Science, Computer Engineering & Applied Computing , 2024

    2024

  38. [45]

    Resqnets: a residual approach for mitigating barren plateaus in quantum neural networks,

    M. Kashif and S. Al-Kuwari, “Resqnets: a residual approach for mitigating barren plateaus in quantum neural networks,” EPJ Quantum Technology , 2024

    2024

  39. [46]

    Advqunn: A methodology for analyzing the adversarial robustness of quanvolutional neural networks,

    W. El Maouaki, A. Marchisio, T. Said, M. Bennai, and M. Shafique, “Advqunn: A methodology for analyzing the adversarial robustness of quanvolutional neural networks,” in International Conference on Quantum Software (QSW), 2024

    2024

  40. [47]

    Qfal: Quantum federated adversarial learning,

    W. El Maouaki et al., “Qfal: Quantum federated adversarial learning,”arXiv preprint arXiv:2502.21171, 2025

    2025 arXiv

  41. [49]

    E. T. Escudero et al. , Assessing the Impact of Noise on Quantum Neural Networks: An Experimental Analysis . Springer Nature, 2023

    2023

  42. [50]

    Quantum neural networks: A comparative analysis and noise robustness evaluation,

    T. Ahmed et al. , “Quantum neural networks: A comparative analysis and noise robustness evaluation,” arXiv preprint arXiv:2501.14412 , 2025

    2025 arXiv

  43. [51]

    Hqnet: Harnessing quantum noise for effective training of quantum neural networks in nisq era,

    M. Kashif and M. Shafique, “Hqnet: Harnessing quantum noise for effective training of quantum neural networks in nisq era,” arXiv:2402.08475, 2024

    2024 arXiv

  44. [52]

    Robqunns: A methodology for robust quanvolutional neural networks against adversarial attacks,

    W. El Maouaki et al., “Robqunns: A methodology for robust quanvolutional neural networks against adversarial attacks,” in International Conference on Image Processing Challenges and Workshops (ICIPCW) , 2024

    2024

  45. [53]

    Quanvolutional neural networks: Powering image recognition with quantum circuits,

    M. Henderson et al. , “Quanvolutional neural networks: Powering image recognition with quantum circuits,” CoRR, vol. abs/1904.04767, 2019

    1904 arXiv

  46. [54]

    Resqunns:towards enabling deep learning in quantum convolution neural networks,

    M. Kashif and M. Shafique, “Resqunns:towards enabling deep learning in quantum convolution neural networks,” arXiv:2402.09146, 2024

    2024 arXiv

  47. [55]

    Quantum convolutional neural networks,

    I. Cong, S. Choi, and M. D. Lukin, “Quantum convolutional neural networks,” Nature Physics, 2019

    2019

  48. [56]

    The mnist database of handwritten digit images for machine learning research [best of the web],

    L. Deng, “The mnist database of handwritten digit images for machine learning research [best of the web],” IEEE Sig. Proc. Mag. , 2012

    2012

  49. [57]

    Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms,

    H. Xiao, K. Rasul, and R. V ollgraf, “Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms,” 2017. 10

    2017

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.