Pith. sign in

REVIEW 4 major objections 6 minor 35 references

Noise-resistant adaptive Hamiltonian learning

T0 review · 4 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read This paper claims that adding a penalty Hamiltonian to a parameterized quantum circuit makes quantum neural networks noise-resistant, with 98.00% classification accuracy under amplitude damping versus 83.33% for a standard quantum neural…

desk verdict A plausible noise-resistance idea undercut by an invalid operator decomposition and uncontrolled experiments. read the letter →

arxiv 2501.08017 v1 pith:VJPHB3D3 submitted 2025-01-14 quant-ph

Wenxuan Wang This is my paper · ORCID
classification quant-ph MSC 81P68 PACS 03.67.Lx
keywords adaptiveHamiltonianlearningnoise-resistantquantumneuralnetworkparameterizedcircuitmachineamplitudedampingnoisepenaltyNISQdevices
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 introduces adaptive Hamiltonian learning (AHL), a parameterized quantum circuit whose unitary layers include a penalty Hamiltonian in addition to the target and overlap Hamiltonians, and claims that a quantum neural network built from this circuit (the RQNN) becomes noise-resistant on NISQ devices. The headline experimental claim is that under amplitude-damping noise the RQNN classifies nonlinear discrete data with 98.00% training accuracy and 98.67% test accuracy, while the standard QNN baseline reaches 83.33% on the test set, and the RQNN also reproduces cosine and damped-sine curves where the QNN deviates. If correct, the AHL circuit structure itself, not just extra trainable parameters, is what suppresses noise, offering a practical noise-mitigation route that avoids full quantum error correction.

What carries the argument

The central object is the AHL parameterized quantum circuit: a layered product of unitaries generated by $H_b$, $H_{\mathrm{olap}}$, and $H_{\mathrm{redun}}$, whose final factor comes from the penalty Hamiltonian $H_{\mathrm{redun}} = \sum \hbar \sigma_n^x$. This term is the mechanism that is supposed to give noise resistance; the paper asserts it actively steers a disturbed state back toward the previous state during evolution, and the whole circuit is decomposed into Pauli rotation X and Z gates plus CNOT gates, making it implementable on current hardware. The parameterized angles $\theta$, $\rho$, and $\gamma$ are what let the RQNN be trained as a neural network by gradient descent.

What would settle it

Run the RQNN classification task under the same amplitude-damping noise with the penalty-Hamiltonian unitary replaced by an identity operation, or by a plain Pauli-X rotation with the same number of angles but no penalty-Hamiltonian interpretation, keeping all other circuit layers and hyperparameters fixed; if accuracy stays near 98%, the claim that the penalty Hamiltonian supplies the noise resistance is falsified. A complementary check is to measure the overlap between noisy and noiseless states at intermediate circuit depths, with and without the penalty term, to test whether it actually moves a noisy state back toward the noiseless evolved state.

Watch

Extended reading notes

Core claim

The paper sets out to establish that decomposing the unitary product $U(\theta,\rho,\gamma) = \prod_{\alpha} e^{-iH_b \theta_\alpha} e^{-iH_{\mathrm{olap}} \rho_\alpha} e^{-iH_{\mathrm{redun}} \gamma_\alpha}$ into elementary gates produces a quantum circuit whose evolution corrects for noise. The final Hamiltonian $H_p$ contains the overlap Hamiltonian $H_{\mathrm{olap}}$ and the penalty Hamiltonian $H_{\mathrm{redun}} = \sum \hbar \sigma_n^x$, and the paper states that when noise interferes during evolution, the penalty term adaptively rectifies the present quantum state back to the prior quantum state. Training the resulting RQNN by finite-difference gradient updates gives accurate function simulation and 98.00% classification accuracy under amplitude-damping noise, compared with 83.33% for a quantum neural network without this circuit, which the paper takes as evidence that the AHL ansatz is noise-resistant on NISQ devices.

Load-bearing premise

The paper assumes, without derivation or error analysis, that adding the penalty-Hamiltonian term to the circuit makes a noisy quantum state return to the ideal state during evolution, and that this correction, rather than extra trainable rotations or a different starting point, is what produces the higher accuracy.

Editorial extensions

If this is right

  • Under amplitude-damping noise, the RQNN maintains high classification accuracy (98% training, 98.67% test) where a standard QNN drops to 83.33% test accuracy, so the AHL ansatz can serve as a drop-in neuron replacement for robust QNNs.
  • The same circuit reproduces continuous target functions (cosine and damped sine) under noise, meaning AHL-trained circuits can be used for data fitting on NISQ hardware.
  • Because the circuit is built only from Rx, Rz, and CNOT gates, the construction transfers to quantum machines that support those elementary gates without additional error-correction overhead.
  • Since function-simulation accuracy improves with circuit depth, the AHL construction gives a controlled way to trade circuit size against noise robustness.

Reading between the lines

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

  • The reported gain could come from the extra parameterized rotations in the RQNN ansatz rather than from the penalty Hamiltonian; a matched-parameter comparison with the same number of variational angles but without the penalty factor would separate those explanations.
  • If the correction mechanism is genuine, it should also suppress dephasing or bit-flip noise at least partially, because the penalty term is a transverse-field term; testing RQNN under depolarizing and dephasing channels would map the scope of the claimed robustness.
  • The mechanism suggests an analogy with symmetry-protected topological order, where noise resistance comes from the ground-state structure; varying the lattice size or boundary conditions and checking whether robustness disappears with trivial topology would probe that analogy.
  • A scaling test with more than four qubits and stronger noise rates would show whether the effect is a small-circuit artefact or a general property of the ansatz on larger NISQ devices.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes an Adaptive Hamiltonian Learning (AHL) model whose parameterized quantum circuit is built from unitaries generated by an initial Hamiltonian H_b, an 'olap' Hamiltonian H_olap, and a penalty Hamiltonian H_redun = Σ ħσ_n^x. The same circuit is used as the neuron of a noise-resistant quantum neural network (RQNN), and the authors claim that the penalty Hamiltonian gives the circuit an adaptive noise-rectifying property. Experiments are reported in the Paddle Quantum framework for mathematical-function simulation and for classification of non-linear discrete data under amplitude damping noise, with a headline test accuracy of 98.00% for RQNN versus 83.33% for a standard QNN. The paper does not provide a derivation of the noise-resistance mechanism, the circuit decomposition in Eq. (17) is inconsistent with the circuit shown in Fig. 3, and the experimental results are reported as single-run accuracies without error bars or statistical analysis.

Significance. If established, a provably noise-resistant parameterized quantum circuit for Hamiltonian learning would be a useful contribution to NISQ-era quantum machine learning. The paper identifies an important problem and offers a concrete ansatz plus comparative experiments, and the loss-curve and function-approximation figures are clearly presented. However, the central theoretical mechanism is not demonstrated: the claimed equivalence between the AHL Hamiltonian and the implemented circuit fails, and the noise-resistance property is asserted rather than derived. The empirical evidence is too weak to support the headline claim, because the accuracy comparison is based on single runs of circuits that differ in structure and parameter count. The paper does not ship machine-checked proofs, reproducible code, or parameter-free derivations, so its value currently rests on an unverified qualitative idea.

major comments (4)
  1. [Eqs. (16)-(17), Fig. 3] The claimed equivalence between the AHL Hamiltonian and the Fig. 3 circuit is not valid. Equation (17) factors exp[-i Σ_{j=k=1}^S (π/2) J_{jk}(σ_j^z + σ_k^x) ρ] into a product of exp[-i(π/2)J_{jk}σ_j^z ρ] and exp[-i(π/2)J_{jk}σ_k^x ρ]. This equality holds only if every σ_j^z commutes with every σ_k^x. Under the paper's description of a spin-1/2 system on lattice edges, σ_j^z and σ_j^x are Pauli operators on the same qubit for j=k and anticommute, so the factorization is false and a correct implementation would require an Euler decomposition rather than independent rotations. If, alternatively, σ_j^z and σ_k^x act on disjoint registers, then H_olap in Eq. (10) is a sum of commuting single-qubit terms and cannot generate the CNOT gates shown in Fig. 3 and mentioned in the text. Either way, the unitary actually implemented on the device is not the unitary on which the noise-resistance argument and the 98% accuracy claim are based.
  2. [Sec. 3.1, after Eq. (11)] The central noise-resistance mechanism is asserted, not derived. The statement that H_redun = Σ ħσ_n^x will 'adaptively rectify the present quantum state to the prior quantum state' under amplitude damping is not supported by a master-equation, Lindblad, or channel-composition analysis. Amplitude damping is a non-unitary channel, and adding a time-independent Pauli-X term to the Hamiltonian does not by itself invert or suppress that channel during unmonitored evolution. Because this assertion underlies the paper's title and main claim, the claimed noise-resistance property of RQNN is currently without theoretical support.
  3. [Sec. 5.2, Table 2] The headline accuracy comparison (RQNN 0.9867 versus QNN 0.8333) is reported for single runs with no error bars, no number of random seeds, no confidence intervals, and no statistical test. Since the RQNN and QNN circuits differ in structure (Figs. 5 and 10) and in parameter count, the observed gap cannot be attributed to the penalty Hamiltonian or to noise resistance; it could reflect expressivity, initialization, or optimization differences. The classification experiment also omits the amplitude-damping rate and the exact data-encoding procedure, so the result is not reproducible as reported.
  4. [Sec. 3.1, Eqs. (2) and (9)] The paper invokes adiabatic evolution and topological protection as the conceptual basis for AHL, but the circuit in Fig. 3 is a fixed sequence of R_x, R_z, and CNOT gates with trained angles; there is no adiabatic schedule s(t) and no adiabatic-theorem argument connecting H_b to H_p. The topological-system language is therefore not substantiated and does not contribute to the noise-resistance claim in any concrete way.
minor comments (6)
  1. [Eq. (9)] The notation Σ_{j=k=1}^S is nonstandard and should be defined explicitly; as written, it suggests j and k are locked together, which creates the commutation ambiguity discussed above.
  2. [Eqs. (9)-(11)] The Planck constant ħ in H_redun plays the role of a coupling constant, and the parameters θ, ρ, γ are introduced without units, leaving the exponents in Eq. (16) dimensionally ambiguous.
  3. [Eqs. (18)-(19)] The loss-function presentation is confusing: y denotes both the label and the output, and Eq. (19) y(i) = (θ,ρ,γ)x(i) + ε(i) treats the circuit parameters as if they were linear regression coefficients, which is inconsistent with the nonlinear parameterized circuit described elsewhere.
  4. [Algorithm 1] The pseudocode contains undefined quantities (N_t, N_s, M, R, E_t, g, ξ) and an incomplete loop header 'For j < ξ'; these should be fully specified.
  5. [References] References [12] and [33] are the same paper (J. Wang et al., Nature Physics 13, 551-555, 2017); one of the duplicate entries should be removed.
  6. [Sec. 5.1, Fig. 6] The noise-free comparison in Fig. 6(d) is not described in the experimental protocol, and no amplitude-damping rate or noise model parameters are given for the noisy simulations, so the comparison between noisy and noise-free settings is not quantitative.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the noise-resistance claim is an assumption tested on held-out data, and the only self-citation is background.

full rationale

The paper's central claim is that the adaptive Hamiltonian learning (AHL) circuit constructed from Hb, Holap, and Hredun (Eqs. (8)-(16)) produces noise-resistant quantum neural networks. The specific noise-rectification property of Hredun is asserted in Section 3.1 ('If noise interference occurs during the evolution process, the penalty Hamiltonian Hredun in Eq. (11) will allow PQC to adaptively rectify the present quantum state to the prior quantum state') rather than derived, but an asserted mechanism validated by experiment is not a circular argument. The headline accuracy of 98.00% is reported as testing-dataset accuracy (Table 2: 0.9867), and the QNN baseline is a separate circuit evaluated under the same amplitude-damping noise, so the result is an independent empirical comparison, not a fitted parameter renamed as a prediction. The only apparent self-citation is Ref. [1] (Shi, W. Wang, et al.), used as background on parameterized Hamiltonian learning; no load-bearing premise, uniqueness theorem, or ansatz is imported from the author's prior work. The factorization in Eq. (17) is mathematically suspect because sigma_z and sigma_x on the same qubit do not commute, so the circuit may not implement the advertised Hamiltonian; however, that is a correctness or validity concern, not a circularity, because the claimed equivalence is not true by construction. Accordingly, no circular step is identified and the paper is scored 1 only for the minor non-load-bearing self-citation.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The central claims rest on these unproven assumptions: the noise-correction property of H_redun, the validity of the factorized decomposition, and the relevance of topological protection to the trained circuit. The fitted circuit parameters are trained on the data, so the 98% accuracy is not an independent prediction.

free parameters (3)
  • θ, ρ, γ (per-layer rotation angles in AHL circuit) = not reported (trained per task)
    Updated via finite-difference gradient descent (Eqs. 20-22); the reported accuracy is the performance of the fitted model.
  • J_ik (coupling strength in H_p) = not specified in paper
    Appears in Eq. (9) and in the Rz/Rx decompositions in Eq. (17); must be chosen by the implementer and affects the circuit.
  • V_n (nuclear spin shift in H_b) = not specified in paper
    Appears in Eq. (8) and in the Rx decompositions; an unstated hyperparameter.
assumptions (3)
  • domain assumption Adiabatic and topological protection properties of the initial Hamiltonian H_b transfer to the trained AHL circuit.
    Invoked in Sections 2.1 and 3.1; no proof shows the trained circuit inherits topological noise protection.
  • ad hoc to paper The exponential of a sum of non-commuting Pauli terms can be factored into the product of individual rotations.
    Eq. (17) separates e^{-i(H_z + H_x)ρ} into e^{-iH_zρ}e^{-iH_xρ}; this requires the terms to commute, which the paper does not establish for same-qubit σ_z and σ_x.
  • ad hoc to paper The penalty Hamiltonian H_redun actively restores the quantum state after noise.
    Asserted in Section 3.1 after Eq. (11) without derivation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Noise-resistant adaptive Hamiltonian learning." pith.science (2026). https://pith.science/paper/VJPHB3D3

@misc{pith2026250108017,
  author       = {Pith},
  title        = {Pith review of: Noise-resistant adaptive Hamiltonian learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VJPHB3D3}},
  note         = {Machine review of arXiv:2501.08017}
}
abstract

Mitigating and reducing noise influence is crucial for obtaining precise experimental results from noisy intermediate-scale quantum (NISQ) devices. In this work, an adaptive Hamiltonian learning (AHL) model for data analysis and quantum state simulation is proposed to overcome problems such as low efficiency and the noise influence of quantum machine learning algorithms. First, an adaptive parameterized quantum circuit with noise resistant ability is constructed by decomposing the unitary operators that include penalty Hamiltonian in the topological quantum system. Then, a noise-resistant quantum neural network (RQNN) based on AHL is developed, which improves the noise robustness of the quantum neural network by updating iterative parameters. Finally, the experiments on Paddle Quantum demonstrate that RQNN can simulate the mathematical function and get accurate classification results on NISQ devices. Compared with the quantum neural network, RQNN ensures high accuracy with the same non-linear discrete data classification under the impact of amplitude damping noise, with an accuracy of 98.00 $\%$. It provides new possibilities for solving practical issues on NISQ devices and also benefits in the resolution of increasingly complicated problems, which will expand the range of potential applications for quantum machine learning models in the future.

Figures

Figures reproduced from arXiv: 2501.08017 by the authors.

Figure 1
Figure 1. Hamiltonian learning model. The quantum state [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. The adaptive Hamiltonian learning model. We design the topological quantum system [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. The parameterized quantum circuit of AHL.The circuit comprises [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: The network structure of RQNN. Rx(θ) Rz(ρ) • Rx(γ) Rx(θ) Rx(ρ) Rx(γ) (a) The Neuron of RQNN. Rx(α) Rz(θ) • Rx(α) Rx(θ) (b) The Neuron of QNN [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: The network structure of QNN and RQNN for simulating function. [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: (a),(b),(c),(d) are the experiment results with RQNN for using data to simulate the function in [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: The Experiment results with RQNN and QNN for training data to the cosine function and the results for RQNN in unnoisy . [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: The Experiment results with RQNN and QNN for training data to simulate the damped sine wave function. [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]
Figure 9
Figure 9. Figure 9: The loss function with RQNN and QNN. TABLE 3 The Comparison of Quantum Neural Networks that can mitigate noise influence Standard Behrman [29] Krisnanda [30] Patterson [31] Konar [32] RQNN General Character Mitigate Noise √ √ √ √ √ Fast Convergent Loss Function √ √ √ √…
Figure 10
Figure 10. Figure 10: The network structure of QNN and RQNN for data classification. [PITH_FULL_IMAGE:figures/full_fig_p009_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 32 canonical work pages

  1. [1]

    Parameterized hamiltonian learning with quantum circuit,

    J. Shi, W. Wang, X. Lou, S. Zhang, and X. Li, “Parameterized hamiltonian learning with quantum circuit,” IEEE T ransactions on Pattern Analysis and Machine Intelligence , 2022

    work page 2022

  2. [2]

    Deepeit: deep image prior enabled electrical impedance tomography,

    D. Liu, J. Wang, Q. Shan, D. Smyl, J. Deng, and J. Du, “Deepeit: deep image prior enabled electrical impedance tomography,”IEEE T ransactions on Pattern Analysis and Machine Intelligence , 2023

    work page 2023

  3. [3]

    Recent advances for quantum neural networks in generative learning,

    J. Tian, X. Sun, Y. Du, S. Zhao, Q. Liu, K. Zhang, W. Yi, W. Huang, C. Wang, X. Wu et al. , “Recent advances for quantum neural networks in generative learning,” IEEE T ransactions on Pattern Analysis and Machine Intelligence , 2023

    work page 2023

  4. [4]

    Quantum advantage in learning from experiments,

    H.-Y. Huang, M. Broughton, J. Cotler, S. Chen, J. Li, M. Mohseni, H. Neven, R. Babbush, R. Kueng, J. Preskill et al. , “Quantum advantage in learning from experiments,” Science, vol. 376, no. 6598, p. 1182–1186, 2022

    work page 2022

  5. [5]

    Negation of the quantum mass function for multisource quantum information fusion with its application to pattern classification,

    F. Xiao and W. Pedrycz, “Negation of the quantum mass function for multisource quantum information fusion with its application to pattern classification,” IEEE T ransactions on Pattern Analysis and Machine Intelligence, 2022

    work page 2022

  6. [6]

    Adaptive pruning-based optimization of parameterized quantum circuits,

    S. Sim, J. Romero, J. F. Gonthier, and A. A. Kunitsa, “Adaptive pruning-based optimization of parameterized quantum circuits,” Quantum Science and T echnology, vol. 6, no. 2, p. 025019, 2021

    work page 2021

  7. [7]

    Hamiltonian learning for quantum error correction,

    A. Valenti, E. van Nieuwenburg, S. Huber, and E. Greplova, “Hamiltonian learning for quantum error correction,” Physical Review Research, vol. 1, no. 3, p. 033092, 2019

    work page 2019

  8. [8]

    Quantum hamilto- nian learning using imperfect quantum resources,

    N. Wiebe, C. Granade, C. Ferrie, and D. Cory, “Quantum hamilto- nian learning using imperfect quantum resources,”Physical Review A, vol. 89, no. 4, p. 042314, 2014

    work page 2014

Show all 35 references
  1. [9]

    Hamiltonian learning and certification using quantum resources,

    N. Wiebe, C. Granade, C. Ferrie, and D. G. Cory, “Hamiltonian learning and certification using quantum resources,” Physical Re- view Letters, vol. 112, no. 19, p. 190501, 2014

    2014

  2. [11]

    Characterization and verification of trotterized digital quantum simulation via hamiltonian and liouvillian learning,

    L. Pastori, T. Olsacher, C. Kokail, and P . Zoller, “Characterization and verification of trotterized digital quantum simulation via hamiltonian and liouvillian learning,” PRX Quantum, vol. 3, no. 3, p. 030324, 2022

    2022

  3. [12]

    A hybrid quantum-classical hamiltonian learning algorithm,

    Y. Wang, G. Li, and X. Wang, “A hybrid quantum-classical hamiltonian learning algorithm,” Science China-Information Sci- ences, vol. 66, no. 2, 2023

    2023

  4. [13]

    Parameterized quantum circuits as machine learning models,

    M. Benedetti, E. Lloyd, S. Sack, and M. Fiorentini, “Parameterized quantum circuits as machine learning models,” Quantum Science and T echnology, vol. 4, no. 4, p. 043001, 2019

    2019

  5. [14]

    Resource-efficient quantum algorithm for protein folding,

    A. Robert, P . K. Barkoutsos, S. Woerner, and I. Tavernelli, “Resource-efficient quantum algorithm for protein folding,” npj Quantum Information, vol. 7, no. 1, p. 38, 2021

    2021

  6. [15]

    Quantum machine learn- ing predicting adme-tox properties in drug discovery,

    A. S. Bhatia, M. K. Saggi, and S. Kais, “Quantum machine learn- ing predicting adme-tox properties in drug discovery,” Journal of Chemical Information and Modeling , vol. 63, no. 21, p. 6476–6486, 2023

    2023

  7. [16]

    Hybrid quantum- classical convolutional neural network model for image classifica- tion,

    F. Fan, Y. Shi, T. Guggemos, and X. X. Zhu, “Hybrid quantum- classical convolutional neural network model for image classifica- tion,” IEEE T ransactions on Neural Networks and Learning Systems , 2023

    2023

  8. [17]

    Quantum circuit learning with parameterized boson sampling,

    J. Shi, Y. Tang, Y. Lu, Y. Feng, R. Shi, and S. Zhang, “Quantum circuit learning with parameterized boson sampling,” IEEE T rans- actions on Knowledge and Data Engineering , 2021

    2021

  9. [18]

    Benchmarking neural networks for quantum computations,

    N. H. Nguyen, E. C. Behrman, M. A. Moustafa, and J. E. Steck, “Benchmarking neural networks for quantum computations,” IEEE T ransactions on Neural Networks and Learning Systems , vol. 31, no. 7, p. 2522–2531, 2019

    2019

  10. [19]

    Matrix completion via non-convex relaxation and adaptive correlation learning,

    X. Li, H. Zhang, and R. Zhang, “Matrix completion via non-convex relaxation and adaptive correlation learning,” IEEE T ransactions on Pattern Analysis and Machine Intelligence , vol. 45, no. 2, p. 1981– 1991, 2022

    1981

  11. [20]

    Classification with quantum neural net- works on near term processors,

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

    2018 arXiv

  12. [21]

    Allgemeine eigenwerttheorie hermitescher funk- tional operatoren,

    J. v. Neumann, “Allgemeine eigenwerttheorie hermitescher funk- tional operatoren,” Mathematische Annalen , vol. 102, no. 1, p. 49– 131, 1930

    1930

  13. [22]

    Identification of symmetry-protected topological states on noisy quantum computers,

    D. Azses, R. Haenel, Y. Naveh, R. Raussendorf, E. Sela, and E. G. Dalla Torre, “Identification of symmetry-protected topological states on noisy quantum computers,” Physical Review Letters , vol. 125, no. 12, p. 120502, 2020

    2020

  14. [23]

    Adiabatic quantum computation,

    T. Albash and D. A. Lidar, “Adiabatic quantum computation,” Reviews of Modern Physics , vol. 90, no. 1, p. 015002, 2018

    2018

  15. [24]

    Effect of data encoding on the expressive power of variational quantum-machine-learning models,

    M. Schuld, R. Sweke, and J. J. Meyer, “Effect of data encoding on the expressive power of variational quantum-machine-learning models,” Physical Review A , vol. 103, no. 3, p. 032430, 2021

    2021

  16. [25]

    Searching a high performance feature extractor for text recognition network,

    H. Zhang, Q. Yao, J. T. Kwok, and X. Bai, “Searching a high performance feature extractor for text recognition network,” IEEE T ransactions on Pattern Analysis and Machine Intelligence , vol. 45, no. 5, p. 6231–6246, 2022

    2022

  17. [26]

    Co-embedding of nodes and edges with graph neural networks,

    X. Jiang, R. Zhu, S. Li, and P . Ji, “Co-embedding of nodes and edges with graph neural networks,” IEEE T ransactions on Pattern Analysis and Machine Intelligence , 2020

    2020

  18. [27]

    San: Side adapter network for open-vocabulary semantic segmentation,

    M. Xu, Z. Zhang, F. Wei, H. Hu, and X. Bai, “San: Side adapter network for open-vocabulary semantic segmentation,” IEEE T rans- actions on Pattern Analysis and Machine Intelligence , 2023

    2023

  19. [28]

    Difference-based image noise modeling using skellam distribution,

    Y. Hwang, J.-S. Kim, and I. S. Kweon, “Difference-based image noise modeling using skellam distribution,” IEEE T ransactions on Pattern Analysis and Machine Intelligence , vol. 34, no. 7, p. 1329– 1341, 2011

    2011

  20. [29]

    Quantum learning with noise and decoherence: a robust quantum neural net- work,

    E. Behrman, N. Nguyen, and J. Steck, “Quantum learning with noise and decoherence: a robust quantum neural net- work,”Quantum Machine Intelligence , vol. 2, p. 1-15, 2020

    2020

  21. [30]

    Creating and concentrating quantum resource states in noisy environments using a quantum neural network,

    T. Krisnanda, S. Ghosh, T. Paterek, and T. C. Liew, “Creating and concentrating quantum resource states in noisy environments using a quantum neural network,” Neural Networks , vol. 136, p. 141–151, 2021

    2021

  22. [31]

    Quantum state discrimination using noisy quantum neural networks,

    A. Patterson, H. Chen, L. Wossnig, S. Severini, D. Browne, and I. Rungger, “Quantum state discrimination using noisy quantum neural networks,” Physical Review Research, vol. 3, no. 1, p. 013063, 2021

    2021

  23. [32]

    A shallow hybrid classical–quantum spiking feedforward neural network for noise-robust image classification,

    D. Konar, A. D. Sarma, S. Bhandary, S. Bhattacharyya, A. Cangi, and V . Aggarwal, “A shallow hybrid classical–quantum spiking feedforward neural network for noise-robust image classification,” Applied Soft Computing , vol. 136, p. 110099, 2023

    2023

  24. [33]

    Experimental quantum hamiltonian learning,

    J. Wang, S. Paesani, R. Santagati, S. Knauer, A. A. Gentile, N. Wiebe, M. Petruzzella, J. L. O’brien, J. G. Rarity, A. Laing et al., “Experimental quantum hamiltonian learning,” Nature Physics , vol. 13, no. 6, p. 551–555, 2017

    2017

  25. [34]

    Learning a local hamilto- nian from local measurements,

    E. Bairey, I. Arad, and N. H. Lindner, “Learning a local hamilto- nian from local measurements,” Physical Review Letters , vol. 122, no. 2, p. 020504, 2019

    2019

  26. [35]

    Unsupervised graph embedding via adaptive graph learning,

    R. Zhang, Y. Zhang, C. Lu, and X. Li, “Unsupervised graph embedding via adaptive graph learning,” IEEE T ransactions on Pattern Analysis and Machine Intelligence , vol. 45, no. 4, p. 5329– 5336, 2022

    2022

  27. [36]

    Multiview clustering: A scalable and parameter-free bipartite graph fusion method,

    X. Li, H. Zhang, R. Wang, and F. Nie, “Multiview clustering: A scalable and parameter-free bipartite graph fusion method,” IEEE T ransactions on Pattern Analysis and Machine Intelligence , vol. 44, no. 1, p. 330–344, 2020. JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2...

    2020

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.