Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

Quantum automated learning with provable and explainable trainability

T0 review · 3 major / 5 minor · reviewed 2026-08-08 · deepseek-v4-flash

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

Pith's one-line read This paper introduces quantum automated learning, a gradient-free scheme in which data-encoded unitaries and label-guided perturbations drive a quantum state to the global minimum of the training loss, and proves exponential convergence…

desk verdict Serious, mostly rigorous proposal for gradient-free quantum learning; the heavy-tail assumption is the main unproven hinge, and step-complexity could be exponential for positive ground energy. read the letter →

arxiv 2502.05264 v1 pith:M775XJ2V submitted 2025-02-07 quant-ph cs.AIcs.LG

Qi Ye , Shuangyue Geng , Zizhao Han , Weikang Li , L.-M. Duan , Dong-Ling Deng This is my paper · ORCID
classification quant-phcs.AIcs.LG MSC 81P6868Q1268T05 PACS 03.67.Lx
keywords quantumautomatedlearninggradient-freemachineimaginarytimeevolutiontrainabilitygeneralizationboundheavy-tailedHamiltonianpost-selectionstatepreparation
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 proposes quantum automated learning (QAL), a way to train a quantum classifier with no variational parameters and no gradient computation. Training is recast as quantum state preparation: a random initial state is repeatedly acted on by unitaries that encode data samples, with a label-dependent perturbation that suppresses wrong predictions. Averaged over samples, the update is exactly imaginary time evolution under the data-averaged Hamiltonian, so the loss—the energy of the state—decays exponentially to its global minimum. The paper proves a generalization bound of order $\sqrt{\log D/N}$ and demonstrates on images and quantum many-body data that the protocol reaches near-perfect accuracy with constant post-selection success probability. If correct, this sidesteps the three main scaling obstacles of variational quantum machine learning: local minima, barren plateaus, and costly per-parameter gradient estimation.

What carries the argument

The load-bearing object is the non-unitary update $(I-\eta H_x)|\psi\rangle/\|(I-\eta H_x)|\psi\rangle\|$, physically realized by sandwiching a block-encoded label perturbation $M_y$ between $U(x)$ and $U(x)^\dagger$ and post-selecting on the ancilla. Averaged over the training set, one step becomes imaginary time evolution under $H_S$, so the training trajectory is a cooling process whose fixed point is the ground state—the global minimum of the loss. The heavy-tail assumption (Definition S2) is what turns exponential-in-$\beta$ convergence into convergence with constant success probability, and the quadratic form of the loss is what yields the logarithmic-dimension generalization bound through matrix concentration.

What would settle it

Construct a dataset with no heavy tail—for instance, random labels attached to a generic state-encoding circuit, or a random-matrix-like Hamiltonian whose spectrum concentrates in the middle—and run QAL from a maximally mixed state. If the post-selection success probability decays exponentially with the number of steps while the conditional loss stays bounded away from the ground energy, then Theorem 2's practical claim fails for that dataset; Theorem 1 alone would still hold but with negligible success probability.

Watch

Extended reading notes

Core claim

The central claim is that a supervised learning task can be solved by preparing the ground state of the data-averaged Hamiltonian $H_S = \mathbb{E}_{x\sim S} H_x$, where $H_x = I - U(x)^\dagger \Pi_{y(x)} U(x)$. One training step applies $U(x)$, then the perturbation $M_y = |y\rangle\langle y| + (1-\eta)(I - |y\rangle\langle y|)$, then $U(x)^\dagger$, which updates the state as $|\psi\rangle \to (I-\eta H_x)|\psi\rangle / \|(I-\eta H_x)|\psi\rangle\|$. At the ensemble level this is $\rho \to e^{-\eta H_S}\rho e^{-\eta H_S} + O(\eta^2)$, i.e. imaginary time evolution. Theorem 1 proves that for any small constant $c$ one can choose the learning rate and number of steps so that the averaged final loss is at most the ground energy $E_g$ plus $c$, with the excess decaying like $e^{-2\beta\delta}$. Theorem 2 proves that when $H_S$ has a heavy-tailed spectrum—a constant fraction of eigenstates near $E_g$—the post-selection success probability stays constant while the loss becomes near-optimal. Theorem 3 bounds the generalization gap by $\sqrt{4\ln(2^{n+1}/\delta)/N}$ with probability at least $1-\delta$.

Load-bearing premise

The practical guarantee that training succeeds with constant probability rests on the heavy-tail assumption: the data-averaged Hamiltonian must have a constant fraction of eigenstates with energy close to its ground energy, and the paper only verifies this numerically on the datasets it tests rather than proving it for general datasets.

Editorial extensions

If this is right

  • Training a QAL model is provably free of local minima and barren-plateau obstructions, because the loss is quadratic and no variational parameters enter the circuit.
  • On near-term hardware the protocol needs only shallow data-encoding circuits and about $\mathcal{O}(\log k)$-gate perturbation unitaries; the simulations reach about 0.99 accuracy on Fashion MNIST with roughly 290 post-selected runs.
  • The generalization bound means a training set of size $N = \Omega(n)$ suffices to control the gap between training and true loss, so sample complexity grows only logarithmically in the Hilbert-space dimension.
  • The trained state can be reused: gentle label measurements do not destroy it, a few additional training steps recover high accuracy, and shadow tomography can supply the few copies needed for many inference queries.
  • The same protocol handles classical images, Hamiltonian data, and quantum state data by choosing the appropriate unitary encoding for each data type.

Reading between the lines

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

  • Our inference: the heavy-tail assumption is checkable classically before any quantum run—compute the empirical spectrum of $H_S$ from the training set—so one could filter datasets that are predictably bad for QAL.
  • Our inference: the imaginary-time-equivalence picture suggests that decoherence acts like a finite-temperature heat bath; QAL's dissipation may therefore be self-correcting under realistic noise, a property the paper demonstrates numerically but does not prove.
  • Our inference: replacing the label projector $\Pi_y$ with task-dependent reward operators could extend the same automated-cooling mechanism to reinforcement or unsupervised learning, but convergence in those settings is not established here.
  • Our inference: since representation power is delegated to the encoding, testing encodings with provably universal feature maps is the natural next step; without such a result QAL's expressivity is only as good as the chosen circuit family.
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

3 major / 5 minor

Summary. The paper introduces quantum automated learning (QAL), a supervised quantum classification protocol without variational parameters. Training data x are encoded into unitaries U(x), and a random state is repeatedly updated by the non-unitary map U(x)^\dagger M_y U(x), where M_y is a label-dependent perturbation implemented by block encoding and post-selection. In the unnormalized density-matrix formalism the update is (I-\eta H_x)\rho(I-\eta H_x), and averaging over the training set gives an imaginary-time evolution under H_S = E_{x\sim S} H_x. The central analytic results are: (i) Theorem 1 / Theorem S2, exponential convergence of the averaged conditional loss to the ground energy of H_S provided the initial state has nonzero ground-state overlap; (ii) Theorem 2 / Theorems S3-S4, constant post-selection success probability with near-optimal loss under a 'heavy-tailed Hamiltonian' assumption; and (iii) Theorem 3 / Theorem S5, a generalization gap bound of order \sqrt{n/N} obtained from matrix Bernstein inequalities. The paper also reports numerical simulations on Fashion MNIST, MNIST, Aubry-Andr\'e Hamiltonians, and cluster-Ising ground states, together with noise-robustness and state-reusability experiments.

Significance. If the central claims are correct, QAL is a genuinely different training paradigm: it avoids variational parameters and gradient evaluation, and the convergence to the global minimum of the empirical risk is proved rather than assumed. The connection to imaginary time evolution gives a transparent physical picture, and the quadratic form of the loss yields a clean generalization bound. The paper is also careful to state several limitations, such as the open question of universal representation power. The numerical demonstrations on several datasets and the explicit noise-robustness results are valuable. However, the practical significance of the main theorem is substantially weakened by the issues below, especially the gap in the proof of Theorem S3 and the unsupported scaling of the heavy-tail assumption. The paper deserves serious consideration, but it needs major revision before the central practical claims can be accepted.

major comments (3)
  1. [Supplementary Sec. III C (Theorem S3), Eq. (S27)] The equality in Eq. (S27) is algebraically incorrect. With beta chosen as beta = 3 ln(1+c2)/(c3 epsilon), the term ln(1/c2)/(2 beta) equals c3 epsilon ln(1/c2)/(6 ln(1+c2)), which is strictly larger than c3 epsilon/6 for every c2 in (0,1/10). Therefore the displayed bound leading to 'g + epsilon + c3' does not follow from the preceding inequalities. Since Theorem S4 and hence Theorem 2 rely directly on Theorem S3, the proof of the constant-success-probability claim is incomplete. The argument is likely repairable by taking beta proportional to ln(1/c2)/(c3 epsilon) rather than ln(1+c2)/(c3 epsilon), but the theorem statement and its constants need to be reworked.
  2. [Definition S2 and Theorem S4] The heavy-tail assumption (Definition S2) is the only bridge between generic convergence in expectation and the constant post-selection success probability claimed in Theorem 2. The assumption is not derived from any data model; the heuristic in the main text about dogs and cats as a mixture of two random projectors is an illustration, not a proof. The numerical support in Fig. 2d and Figs. S3-S5 is restricted to ten-qubit datasets and a fixed accuracy parameter, and no evidence is given that the low-energy eigenstate fraction c2 stays constant as n grows with the data dimension or as epsilon shrinks toward the target accuracy. If c2 decays with n or epsilon, the success probability lower bound in Theorem S3 decays correspondingly and the resource cost becomes exponential. The authors should either prove heavy-tailedness for a concrete class of data-encoding Hamiltonians or explicitly state this as a limitation and provide scaling experiments.
  3. [Supplementary Sec. III B (Theorem S2)] Theorem S2 is existential and the proof gives no bound on the required number of steps T. From the proof conditions e^{-2 beta delta} < sigma_g c/4 and beta eta e^{2 beta g} < sigma_g c/16, one needs eta of order sigma_g e^{-2 beta g}/beta, so T = beta/eta grows at least like e^{2 beta g}/sigma_g. For the generic case g > 0 this is exponential in beta g, and beta itself must be about (1/delta) ln(1/(sigma_g c)) to reach error c. Thus the main-text statement that 'the training process will converge exponentially as T increases' does not by itself imply practical trainability. The paper should report the dependence of T on g, delta, and sigma_g, and clarify whether the claim is convergence for sufficiently large T or efficient trainability in the system size.
minor comments (5)
  1. [Main text Eq. (3) and Supplementary Eq. (S28)] The notation '2n+1' should be '2^{n+1}' in the logarithm; as written it does not match the Hilbert-space dimension used in the proof.
  2. [Theorem 2 and Theorem S4] Theorem 2 says 'with a random initial state in the computational basis', but the proof of Theorem S4 uses the maximally mixed initial state I/2^n. The relation should be clarified: the averaged state over random computational-basis inputs is maximally mixed, but a single random pure state is not covered by the stated proof.
  3. [Fig. 2c] The caption states that the success probability is calculated analytically from Eq. (2) with the O(T eta^2) term omitted; for eta = 0.1 and the large step counts shown, T eta^2 is not obviously small, so the plotted trade-off should be validated against the exact unnormalized-state simulation.
  4. [Various] There are minor typographical issues: 'ploted' in Methods, 'Supplimentary' in Supplementary Sec. IV, and reference [57] duplicates reference [50].
  5. [Discussion] The statement that QAL 'escapes the barren plateau problem inherently' is too strong without a complexity analysis of the training circuit; the absence of variational parameters removes gradient-vanishing concerns, but the circuit depth and post-selection overhead have not been analyzed as a function of n.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the QAL convergence proof is a genuine dynamical analysis of the explicitly defined update rule and loss function.

full rationale

The derivation chain is self-contained. The training update is explicitly defined as |ψ> ← (I − ηH_x)|ψ> / ||(I − ηH_x)|ψ>|| (Eq. 1), and the loss is defined as R_S(ψ) = <ψ|H_S|ψ> with H_S = E_{x∼S}H_x. The convergence theorem (Theorem 1 / Theorem S2) analyzes the averaged dynamics of this update, showing that it approximates imaginary time evolution under H_S and that the loss gap decays as e^{−2βδ} plus controlled error terms. This is a mathematical proof about the defined algorithm, not a prediction that merely restates an input. The heavy-tail assumption (Definition S2) is an independent domain assumption about the spectrum of H_S; it is not an output of the derivation, and it is only used to guarantee a constant post-selection success probability in Theorem 2 / Theorem S4. The paper numerically validates this assumption on specific datasets, which is evidence but not a circular step. The generalization bound (Theorem 3 / Theorem S5) follows from matrix Bernstein inequalities applied to H_x, again independent of the training dynamics. Self-citations in the paper (e.g., refs. 6, 19, 42, 44, 47) are contextual tutorial/review references or references to prior algorithmic tools; none of them is load-bearing for the convergence, success-probability, or generalization theorems. No fitted parameter is renamed as a prediction, and no ansatz is smuggled in via self-citation. The main practical caveat—that the heavy-tail assumption is not proven for generic datasets and that the required T can be large when the ground energy is positive—is a correctness/efficiency concern, not a circularity concern.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The central claim rests on standard analytic tools and two domain assumptions: initial overlap with the ground space (mild) and heavy-tailed spectrum of the data Hamiltonian H_S (strong, data-dependent). No new physical entities are introduced.

free parameters (1)
  • learning rate η = 0.1 in simulations
    Hyperparameter set to 0.1 for most simulations; the proof requires choosing η sufficiently small depending on the spectral gap and ground energy, so in practice it would need tuning.
assumptions (4)
  • domain assumption Initial state ρ0 has a nonzero overlap with the ground space of H_S
    Needed for Theorem 1; can be ensured by starting with the maximally mixed state. This is a mild assumption.
  • domain assumption Heavy-tail assumption: H_S has a constant proportion of low-energy eigenstates
    Needed for Theorem 2 to guarantee constant post-selection success probability; not generally true and only validated numerically on the tested datasets.
  • domain assumption Data samples are drawn independently from an unknown distribution D
    Standard i.i.d. assumption needed for the generalization bound.
  • standard math H_x are projectors with eigenvalues 0 and 1
    Follows from the definition H_x = I - U(x)† Π_y U(x); used in matrix Bernstein bound.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Quantum automated learning with provable and explainable trainability." pith.science (2026). https://pith.science/paper/M775XJ2V

@misc{pith2026250205264,
  author       = {Pith},
  title        = {Pith review of: Quantum automated learning with provable and explainable trainability},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/M775XJ2V}},
  note         = {Machine review of arXiv:2502.05264}
}
read the original abstract

Machine learning is widely believed to be one of the most promising practical applications of quantum computing. Existing quantum machine learning schemes typically employ a quantum-classical hybrid approach that relies crucially on gradients of model parameters. Such an approach lacks provable convergence to global minima and will become infeasible as quantum learning models scale up. Here, we introduce quantum automated learning, where no variational parameter is involved and the training process is converted to quantum state preparation. In particular, we encode training data into unitary operations and iteratively evolve a random initial state under these unitaries and their inverses, with a target-oriented perturbation towards higher prediction accuracy sandwiched in between. Under reasonable assumptions, we rigorously prove that the evolution converges exponentially to the desired state corresponding to the global minimum of the loss function. We show that such a training process can be understood from the perspective of preparing quantum states by imaginary time evolution, where the data-encoded unitaries together with target-oriented perturbations would train the quantum learning model in an automated fashion. We further prove that the quantum automated learning paradigm features good generalization ability with the generalization error upper bounded by the ratio between a logarithmic function of the Hilbert space dimension and the number of training samples. In addition, we carry out extensive numerical simulations on real-life images and quantum data to demonstrate the effectiveness of our approach and validate the assumptions. Our results establish an unconventional quantum learning strategy that is gradient-free with provable and explainable trainability, which would be crucial for large-scale practical applications of quantum computing in machine learning scenarios.

Figures

Figures reproduced from arXiv: 2502.05264 by the authors.

Figure 1
Figure 1. FIG. 1 [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2 [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Stochastic Quantum Hamiltonian Descent

    quant-ph 2025-07 conditional novelty 6.0 of 10

    SQHD is a gate-based quantum algorithm that approximates a Lindblad dynamics blending Hamiltonian descent with stochastic component noise, giving an order-2 weak approximation and an O(1/t + eta sigma*) convergence bo...

Reference graph

Works this paper leans on

83 extracted references · 61 canonical work pages · cited by 1 Pith paper

  1. [1]

    LeCun, Y

    Y . LeCun, Y . Bengio, and G. Hinton, Deep learning, Nature 521, 436 (2015)

    2015

  2. [2]

    Goodfellow, Y

    I. Goodfellow, Y . Bengio, and A. Courville, Deep Learning (MIT Press, 2016)

    2016

  3. [3]

    H. Wang, T. Fu, Y . Du,et al., Scientific discovery in the age of artificial intelligence, Nature 620, 47 (2023)

    2023

  4. [4]

    Silver, A

    D. Silver, A. Huang, C. J. Maddison, et al., Mastering the game of Go with deep neural networks and tree search, Nature 529, 484 (2016)

    work page 2016

  5. [5]

    Silver, J

    D. Silver, J. Schrittwieser, K. Simonyan, et al., Mastering the game of Go without human knowledge, Nature550, 354 (2017)

    work page 2017

  6. [6]

    A. W. Senior, R. Evans, J. Jumper, et al. , Improved protein structure prediction using potentials from deep learning, Nature 577, 706 (2020)

    work page 2020

  7. [7]

    T. H. Trinh, Y . Wu, Q. V . Le, H. He, and T. Luong, Solv- ing olympiad geometry without human demonstrations, Nature 625, 476 (2024)

    work page 2024

  8. [8]

    Electron

    AI hardware has an energy problem, Nat. Electron. 6, 463 (2023)

    work page 2023

Show all 83 references
  1. [9]

    Preskill, Quantum Computing in the NISQ era and beyond, Quantum 2, 79 (2018)

    J. Preskill, Quantum Computing in the NISQ era and beyond, Quantum 2, 79 (2018)

    2018

  2. [10]

    Arute, K

    F. Arute, K. Arya, R. Babbush, et al., Quantum supremacy us- ing a programmable superconducting processor, Nature 574, 505 (2019)

    2019

  3. [11]

    Zhong, H

    H.-S. Zhong, H. Wang, Y .-H. Deng, et al., Quantum computa- tional advantage using photons, Science 370, 1460 (2020)

    2020

  4. [12]

    Wu, W.-S

    Y . Wu, W.-S. Bao, S. Cao, et al., Strong Quantum Computa- tional Advantage Using a Superconducting Quantum Processor, Phys. Rev. Lett. 127, 180501 (2021)

    2021

  5. [13]

    Bluvstein, S

    D. Bluvstein, S. J. Evered, A. A. Geim, et al., Logical quantum processor based on reconfigurable atom arrays, Nature 626, 58 (2024)

    2024

  6. [14]

    Acharya, D

    R. Acharya, D. A. Abanin, L. Aghababaie-Beni, I. Aleiner, T. I. Andersen, M. Ansmann, F. Arute, K. Arya, A. Asfaw, N. As- trakhantsev, et al., Quantum error correction below the surface code threshold, Nature , 1 (2024)

    2024

  7. [15]

    Y . Liu, S. Arunachalam, and K. Temme, A rigorous and robust quantum speed-up in supervised machine learning, Nat. Phys. 17, 1013 (2021)

    2021

  8. [16]

    Huang, M

    H.-Y . Huang, M. Broughton, M. Mohseni, R. Babbush, S. Boixo, H. Neven, and J. R. McClean, Power of data in quan- tum machine learning, Nat. Commun. 12, 2631 (2021)

    2021

  9. [17]

    Biamonte, P

    J. Biamonte, P. Wittek, N. Pancotti, P. Rebentrost, N. Wiebe, and S. Lloyd, Quantum machine learning, Nature 549, 195 (2017)

    2017

  10. [18]

    Dunjko and H

    V . Dunjko and H. J. Briegel, Machine learning & artificial in- telligence in the quantum domain: A review of recent progress, Rep. Prog. Phys. 81, 074001 (2018)

    2018

  11. [19]

    Das Sarma, D.-L

    S. Das Sarma, D.-L. Deng, and L.-M. Duan, Machine learning meets quantum physics, Physics Today 72, 48 (2019)

    2019

  12. [20]

    Herrmann, S

    J. Herrmann, S. M. Llima, A. Remm, et al., Realizing quan- tum convolutional neural networks on a superconducting quan- tum processor to recognize quantum phases, Nat. Commun.13, 4144 (2022)

    2022

  13. [21]

    Saggio, B

    V . Saggio, B. E. Asenbeck, A. Hamann, et al. , Experimen- tal quantum speed-up in reinforcement learning agents, Nature 591, 229 (2021)

    2021

  14. [22]

    W. Ren, W. Li, S. Xu, Wang, et al., Experimental quantum ad- versarial learning with programmable superconducting qubits, Nat. Comput. Sci. 2, 711 (2022)

    2022

  15. [23]

    Hu, S.-H

    L. Hu, S.-H. Wu, W. Cai, et al., Quantum generative adversar- ial learning in a superconducting quantum circuit, Sci. Adv. 5, eaav2761 (2019)

    2019

  16. [24]

    Huang, Z.-A

    K. Huang, Z.-A. Wang, C. Song, et al. , Quantum generative adversarial networks with multiple superconducting qubits, npj. Quantum. Inf. 7, 165 (2021)

    2021

  17. [25]

    Gong, H.-L

    M. Gong, H.-L. Huang, S. Wang, et al. , Quantum neu- ronal sensing of quantum many-body states on a 61-qubit 9 programmable superconducting processor, Sci. Bull. 68, 906 (2023)

    2023

  18. [26]

    Peters, J

    E. Peters, J. Caldeira, A. Ho, S. Leichenauer, M. Mohseni, H. Neven, P. Spentzouris, D. Strain, and G. N. Perdue, Machine learning of high dimensional data on a noisy quantum proces- sor, npj. Quantum. Inf. 7, 161 (2021)

    2021

  19. [27]

    Liu, X.-Y

    Z.-P. Liu, X.-Y . Cao, H.-W. Liu, X.-R. Sun, Y . Bao, Y .-S. Lu, H.-L. Yin, and Z.-B. Chen, Practical quantum federated learning and its experimental demonstration, arXiv:2501.12709 (2025)

    2025 arXiv

  20. [28]

    Zhang, Z

    C. Zhang, Z. Lu, L. Zhao, et al., Quantum continual learning on a programmable superconducting processor, arXiv:2409.09729 (2024)

    2024

  21. [30]

    B. J. Wythoff, Backpropagation neural networks: A tutorial, Chemometr. Intell. Lab. Syst. 18, 115 (1993)

    1993

  22. [31]

    K. Beer, D. Bondarenko, T. Farrelly, T. J. Osborne, R. Salz- mann, D. Scheiermann, and R. Wolf, Training deep quantum neural networks, Nat. Commun. 11, 808 (2020)

    2020

  23. [32]

    X. Pan, Z. Lu, W. Wang, Z. Hua, Y . Xu, W. Li, W. Cai, X. Li, H. Wang, Y .-P. Song,et al., Deep quantum neural networks on a superconducting processor, Nat. Commun. 14, 4006 (2023)

    2023

  24. [34]

    Achiam, S

    OpenAI, J. Achiam, S. Adler, et al., GPT-4 Technical Report, arXiv:2303.08774 (2024)

    2024 arXiv

  25. [35]

    N. Du, Y . Huang, A. M. Dai, et al. , GLaM: Efficient Scal- ing of Language Models with Mixture-of-Experts, in Proceed- ings of the 39th International Conference on Machine Learning (PMLR, 2022) pp. 5547–5569

    2022

  26. [36]

    Bittel and M

    L. Bittel and M. Kliesch, Training Variational Quantum Algo- rithms Is NP-Hard, Phys. Rev. Lett. 127, 120502 (2021)

    2021

  27. [41]

    H. Xiao, K. Rasul, and R. V ollgraf, Fashion-mnist: A novel image dataset for benchmarking machine learning algorithms, arXiv:1708.07747 (2017)

    2017 arXiv

  28. [42]

    Li and D.-L

    W. Li and D.-L. Deng, Recent advances for quantum classifiers, Sci. China Phys. Mech. Astron. 65, 220301 (2022)

    2022

  29. [43]

    Cerezo, A

    M. Cerezo, A. Arrasmith, R. Babbush, S. C. Benjamin, S. Endo, K. Fujii, J. R. McClean, K. Mitarai, X. Yuan, L. Cincio, and P. J. Coles, Variational quantum algorithms, Nat. Rev. Phys. 3, 625 (2021)

    2021

  30. [44]

    W. Li, Z. Lu, and D.-L. Deng, Quantum Neural Network Clas- sifiers: A Tutorial, SciPost Phys. Lect. Notes 1, 61 (2022)

    2022

  31. [45]

    Giovannetti, S

    V . Giovannetti, S. Lloyd, and L. Maccone, Quantum Random Access Memory, Phys. Rev. Lett. 100, 160501 (2008)

    2008

  32. [46]

    C. M. Dawson and M. A. Nielsen, The solovay-kitaev algo- rithm, Quantum Info. Comput. 6, 81 (2006)

    2006

  33. [47]

    Zhang, P.-L

    Y .-H. Zhang, P.-L. Zheng, Y . Zhang, and D.-L. Deng, Topolog- ical Quantum Compiling with Reinforcement Learning, Phys. Rev. Lett. 125, 170501 (2020)

    2020

  34. [48]

    W. K. Wootters and W. H. Zurek, A single quantum cannot be cloned, Nature 299, 802 (1982)

    1982

  35. [49]

    D. J. Griffiths and D. F. Schroeter, Introduction to quantum me- chanics (Cambridge university press, 2019)

    2019

  36. [51]

    S. Aaronson, Shadow tomography of quantum states, in Pro- ceedings of the 50th Annual ACM SIGACT Symposium on The- ory of Computing , STOC 2018 (Association for Computing Machinery, New York, NY , USA, 2018) pp. 325–338

    2018

  37. [52]

    Kjaergaard, M

    M. Kjaergaard, M. E. Schwartz, J. Braum ¨uller, P. Krantz, J. I.- J. Wang, S. Gustavsson, and W. D. Oliver, Superconducting Qubits: Current State of Play, Annu. Rev. Condens. Matter Phys. 11, 369 (2020)

    2020

  38. [53]

    T. Goto, Q. H. Tran, and K. Nakajima, Universal Approx- imation Property of Quantum Machine Learning Models in Quantum-Enhanced Feature Spaces, Phys. Rev. Lett. 127, 090506 (2021)

    2021

  39. [54]

    L ´opez-Pastor and F

    V . L ´opez-Pastor and F. Marquardt, Self-Learning Machines Based on Hamiltonian Echo Backpropagation, Phys. Rev. X13, 031020 (2023)

    2023

  40. [55]

    Deng et al

    J. Deng et al. , ImageNet: A large-scale hierarchical image database, in 2009 IEEE Conference on Computer Vision and Pattern Recognition (2009) pp. 248–255

    2009

  41. [56]

    Xiang, W

    L. Xiang, W. Jiang, Z. Bao, et al., Long-lived topological time- crystalline order on a quantum processor, Nat. Commun. 15, 8963 (2024)

    2024

  42. [57]

    F. Jin, S. Jiang, X. Zhu, Z. Bao, F. Shen, K. Wang, Z. Zhu, S. Xu, Z. Song, J. Chen, et al. , Observation of topological prethermal strong zero modes, arXiv:2501.04688 (2025)

    2025 arXiv

  43. [58]

    Acharya, L

    R. Acharya, L. Aghababaie-Beni, I. Aleiner, T. Andersen, M. Ansmann, F. Arute, K. Arya, A. Asfaw, N. Astrakhantsev, J. Atalaya, et al., Quantum error correction below the surface code threshold, Nature (2024)

    2024

  44. [59]

    Iqbal, N

    M. Iqbal, N. Tantivasadakarn, R. Verresen, S. L. Campbell, J. M. Dreiling, C. Figgatt, J. P. Gaebler, J. Johansen, M. Mills, S. A. Moses, et al., Non-abelian topological order and anyons on a trapped-ion processor, Nature 626, 505 (2024)

    2024

  45. [60]

    S. Cao, B. Wu, F. Chen, M. Gong, Y . Wu, Y . Ye, C. Zha, H. Qian, C. Ying, S. Guo, et al. , Generation of genuine en- tanglement up to 51 superconducting qubits, Nature 619, 738 (2023)

    2023

  46. [61]

    A. M. Childs and N. Wiebe, Hamiltonian Simulation Using Lin- ear Combinations of Unitary Operations, QIC 12, 901 (2012)

    2012

  47. [66]

    Aubry and G

    S. Aubry and G. Andr ´e, Analyticity breaking and anderson lo- calization in incommensurate lattices, Ann. Israel Phys. Soc 3, 18 (1980)

    1980

  48. [67]

    W. Son, L. Amico, R. Fazio, A. Hamma, S. Pascazio, and V . Ve- dral, Quantum phase transition between cluster and antiferro- magnetic states, EPL 95, 50001 (2011)

    2011

  49. [68]

    projected

    P. Smacchia, L. Amico, P. Facchi, R. Fazio, G. Florio, S. Pas- cazio, and V . Vedral, Statistical mechanics of the cluster ising model, Phys. Rev. A 84, 022304 (2011). Supplementary Information for: Quantum automated learning with provable and explainable trainability CONTENTS...

    2011 arXiv

  50. [69]

    Goodfellow, Y

    I. Goodfellow, Y . Bengio, and A. Courville,Deep Learning (MIT Press, 2016). 12

    2016

  51. [70]

    Cerezo, A

    M. Cerezo, A. Arrasmith, R. Babbush, S. C. Benjamin, S. Endo, K. Fujii, J. R. McClean, K. Mitarai, X. Yuan, L. Cincio, and P. J. Coles, Variational quantum algorithms, Nat. Rev. Phys.3, 625 (2021)

    2021

  52. [71]

    Mitarai, M

    K. Mitarai, M. Negoro, M. Kitagawa, and K. Fujii, Quantum circuit learning, Phys. Rev. A 98, 032309 (2018)

    2018

  53. [72]

    Schuld, V

    M. Schuld, V . Bergholm, C. Gogolin, J. Izaac, and N. Killoran, Evaluating analytic gradients on quantum hardware, Phys. Rev. A 99, 032331 (2019)

    2019

  54. [73]

    Stokes, J

    J. Stokes, J. Izaac, N. Killoran, and G. Carleo, Quantum Natural Gradient, Quantum 4, 269 (2020)

    2020

  55. [74]

    H. Cai, Q. Ye, and D.-L. Deng, Sample complexity of learning parametric quantum circuits, Quantum Sci. Technol. 7, 025014 (2022)

    2022

  56. [75]

    M. C. Caro, H.-Y . Huang, M. Cerezo, K. Sharma, A. Sornborger, L. Cincio, and P. J. Coles, Generalization in quantum machine learning from few training data, Nat. Commun. 13, 4919 (2022)

    2022

  57. [76]

    You and X

    X. You and X. Wu, Exponentially Many Local Minima in Quantum Neural Networks, inProceedings of the 38th International Conference on Machine Learning (PMLR, 2021) pp. 12144–12155

    2021

  58. [77]

    J. R. McClean, S. Boixo, V . N. Smelyanskiy, R. Babbush, and H. Neven, Barren plateaus in quantum neural network training landscapes, Nat. Commun. 9, 4812 (2018)

    2018

  59. [78]

    Cerezo, A

    M. Cerezo, A. Sone, T. V olkoff, L. Cincio, and P. J. Coles, Cost Function Dependent Barren Plateaus in Shallow Parametrized Quantum Circuits, Nat. Commun. 12, 1791 (2021)

    2021

  60. [79]

    Ortiz Marrero, M

    C. Ortiz Marrero, M. Kieferov ´a, and N. Wiebe, Entanglement-Induced Barren Plateaus, PRX Quantum 2, 040316 (2021)

    2021

  61. [80]

    Holmes, K

    Z. Holmes, K. Sharma, M. Cerezo, and P. J. Coles, Connecting ansatz expressibility to gradient magnitudes and barren plateaus, PRX Quantum 3, 010313 (2022)

    2022

  62. [81]

    S. Wang, E. Fontana, M. Cerezo, K. Sharma, A. Sone, L. Cincio, and P. J. Coles, Noise-induced barren plateaus in variational quantum algorithms, Nat. Commun. 12, 6961 (2021)

    2021

  63. [82]

    Larocca, S

    M. Larocca, S. Thanasilp, S. Wang, K. Sharma, J. Biamonte, P. J. Coles, L. Cincio, J. R. McClean, Z. Holmes, and M. Cerezo, A Review of Barren Plateaus in Variational Quantum Computing, arXiv:2405.00781 (2024)

    2024 arXiv

  64. [83]

    A. M. Childs and N. Wiebe, Hamiltonian Simulation Using Linear Combinations of Unitary Operations, QIC 12, 901 (2012)

    2012

  65. [84]

    I. M. Georgescu, S. Ashhab, and F. Nori, Quantum simulation, Rev. Mod. Phys. 86, 153 (2014)

    2014

  66. [85]

    G. H. Low and I. L. Chuang, Optimal hamiltonian simulation by quantum signal processing, Phys. Rev. Lett. 118, 010501 (2017)

    2017

  67. [86]

    Clinton, J

    L. Clinton, J. Bausch, and T. Cubitt, Hamiltonian simulation algorithms for near-term quantum hardware, Nat. Commun.12, 4989 (2021)

    2021

  68. [87]

    Lloyd, M

    S. Lloyd, M. Mohseni, and P. Rebentrost, Quantum principal component analysis, Nat. Phys. 10, 631 (2014)

    2014

  69. [88]

    R. Vale, T. M. D. Azevedo, I. C. S. Ara ´ujo, I. F. Araujo, and A. J. da Silva, Decomposition of Multi-controlled Special Unitary Single- Qubit Gates, arXiv:2302.06377 (2023)

    2023 arXiv

  70. [89]

    Mohri, A

    M. Mohri, A. Rostamizadeh, and A. Talwalkar, Foundations of machine learning (MIT press, 2018)

    2018

  71. [90]

    J. A. Tropp, An Introduction to Matrix Concentration Inequalities, arXiv:1501.01571 (2015)

    2015 arXiv

  72. [91]

    LeCun, C

    Y . LeCun, C. Cortes, and C. J. C. Burges, The MNIST database of handwritten digits (1998)

    1998

  73. [92]

    Aubry and G

    S. Aubry and G. Andr ´e, Analyticity breaking and anderson localization in incommensurate lattices, Ann. Israel Phys. Soc 3, 18 (1980)

    1980

  74. [93]

    W. Son, L. Amico, R. Fazio, A. Hamma, S. Pascazio, and V . Vedral, Quantum phase transition between cluster and antiferromagnetic states, EPL 95, 50001 (2011)

    2011

  75. [94]

    Smacchia, L

    P. Smacchia, L. Amico, P. Facchi, R. Fazio, G. Florio, S. Pascazio, and V . Vedral, Statistical mechanics of the cluster ising model, Phys. Rev. A 84, 022304 (2011)

    2011

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.