pith. sign in

arxiv: 2606.18515 · v1 · pith:TOSVM6WHnew · submitted 2026-06-16 · 🪐 quant-ph · cs.LG· stat.ML

Exponentially many initializations to avoid barren plateaus

Ankit Kulshrestha , Ricard Puig , Diego Garc\'ia-Mart\'in , Lukasz Cincio , Ilya Safro , Zo\"e Holmes , M. Cerezo This is my paper

Pith reviewed 2026-06-26 23:54 UTC · model grok-4.3

classification 🪐 quant-ph cs.LGstat.ML
keywords barren plateausvariational quantum algorithmsparameter initializationfirst-moment diagnosticloss concentrationquantum circuitsinitialization strategies
0
0 comments X

The pith

A first-moment diagnostic reveals exponentially many inequivalent initializations escape barren plateaus in quantum ansatze.

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

The paper establishes that barren plateaus arise as an average-case effect under naive initialization, but a simple operator-level check on the first moment of the loss shows that many different parameter distributions avoid the fully concentrated fixed point. This diagnostic recovers known schemes such as identity and Gaussian initializations while proving that shifted, biased, and non-symmetric distributions also work and need not be equivalent to one another. A sympathetic reader cares because the numerics indicate these distinct initializations reach different attained minima, converting the single problem of exponential concentration into the task of choosing among multiple trainable pockets. The framework therefore supplies both a test for escape and a way to compare biases across strategies.

Core claim

Our results show that one can generate exponentially many families of inequivalent initialization strategies. The first-moment framework supplies an operator-level diagnostic that determines when an initialization escapes the fully concentrated barren-plateau fixed point and that distinguishes the biases induced by different strategies. Many shifted, biased, and non-symmetric parameter distributions satisfy the diagnostic and therefore avoid concentration, yet these choices are not equivalent; numerics further indicate that first-moment-distinct initializations can converge to different minima.

What carries the argument

The first-moment framework: an operator-level diagnostic that checks whether the expected value of the loss under a given parameter distribution remains non-concentrated, thereby flagging escape from the barren-plateau fixed point.

If this is right

  • Barren-plateau avoidance is highly non-unique; many distinct distributions satisfy the escape condition.
  • Inequivalent first-moment-distinct initializations are not required to produce the same final solution quality.
  • Different escaping initializations can be compared directly by the biases their distributions induce on the loss operator.
  • Avoiding concentration via initialization replaces one scaling problem with the task of selecting among multiple distinct trainable regions.

Where Pith is reading between the lines

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

  • The diagnostic could be applied to rank candidate initializations by how strongly they bias the loss toward particular solution classes.
  • One could test whether the exponentially many families partition the space of possible minima into regions with systematically different properties such as depth or entanglement.
  • Combining the first-moment test with a secondary criterion that favors one pocket over others might restore a unique preferred initialization for a given task.

Load-bearing premise

The first-moment diagnostic is sufficient to decide whether an initialization escapes concentration for the full loss landscape and the entire optimization trajectory.

What would settle it

A numerical experiment in which an initialization strategy that satisfies the first-moment diagnostic nevertheless produces loss values that concentrate to an exponentially small variance across random instances.

Figures

Figures reproduced from arXiv: 2606.18515 by Ankit Kulshrestha, Diego Garc\'ia-Mart\'in, Ilya Safro, Lukasz Cincio, M. Cerezo, Ricard Puig, Zo\"e Holmes.

Figure 1
Figure 1. Figure 1: ). We support this picture both analytically and nu￾merically, showing that different initialization families can lead to different landscape patches and thus to different training dynamics, and different attained minima. This manuscript is organized as follows. In Section II, we present the definitions and tools necessary to follow the results. In Sec. III, we start by presenting two theo￾rems which provi… view at source ↗
Figure 2
Figure 2. Figure 2: Circuit ansatz. We consider a PQC composed of single-qubit Rx and Ry rotations followed by nearest-neighbor CZ entangling gates. The example shown corresponds to n = 6 qubits and L = 3 layers. Theorem 3. Let U(θ) be a PQC composed of general single-qubit gates interleaved with diagonal entangling gates Vl, i.e., U(θ) = Y L l=1 Yn j=1 e −iθ1,j,lσ (j) x /2 e −iθ2,j,lσ (j) z /2 e −iθ3,j,lσ (j) x /2Vl , let O … view at source ↗
Figure 3
Figure 3. Figure 3: Distribution of did(θ) for several initialization families and system sizes. We consider an L = 4 layered circuit (see [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Circuit ansatz for Max-Cut example. We con￾sider a PQC composed of single-qubit Rz, Ry, Rz rotations followed by nearest-neighbor CZ entangling gates. The exam￾ple shown corresponds to n = 6 qubits and L = 2 layers. Given that each Ry can be initialized using two distributions, there are 2 nL choices, which we here schematically show by coloring gates according to which initialization was used. where ∥·∥1 … view at source ↗
Figure 6
Figure 6. Figure 6: Gradient norm distributions for different initialization families and system sizes. We consider an L = 2n layered circuit (see [PITH_FULL_IMAGE:figures/full_fig_p041_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Order statistics of gradient distributions around convergent point. We show the expectation value and the variance (inset) for the values of ∥∇L(θp)∥ in [PITH_FULL_IMAGE:figures/full_fig_p043_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Magnitude of gradient norms across different qubits and layers for three different qubit sizes. We consider system sizes of n = 2, 4, 8 qubits (respectively panels a), b) and c)), a number of layers L = 2n, and initialization strategies Unif[−π, π], Beta(α, β), and N (0, 1 n ) [PITH_FULL_IMAGE:figures/full_fig_p044_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Gradient landscape around some local optimum for different parameter initializations and system sizes. We consider system sizes of n = 2, 4, 8 qubits (respectively panels a), b) and c)), a number of layers L = 2n, and initializations strategies Unif[−π, π], Beta(α, β), and N (0, 1 n ) [PITH_FULL_IMAGE:figures/full_fig_p045_9.png] view at source ↗
read the original abstract

Barren plateaus are stated as an average-case phenomenon: pick an ansatz, initialize it naively, and concentration follows. This has led to the common view that a potential cure for barren plateaus is simply to initialize the parameters more carefully. Here we show that the situation is subtler. We introduce a first-moment framework that gives a simple operator-level diagnostic for when an initialization may escape the fully concentrated barren-plateau fixed point, and for comparing the biases induced by different initialization strategies. Our framework recovers several known initialization schemes such as identity and Gaussian initialization, but also shows that barren-plateau avoidance is highly non-unique. Indeed, many shifted, biased, and non-symmetric parameter distributions can avoid concentration, and these choices need not be equivalent. In fact, our results show that one can generate exponentially many families of inequivalent initialization strategies. Then, our numerics indicate that different first-moment-distinct initializations can lead to different attained minima, suggesting that avoiding barren plateaus via smart initializations can trade the exponential concentration problem for the challenge of selecting the right trainable pocket amongst many options.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

1 major / 1 minor

Summary. The manuscript introduces a first-moment operator-level diagnostic for determining when parameter initializations escape the fully concentrated barren-plateau fixed point in variational quantum algorithms. It recovers known schemes such as identity and Gaussian initialization, shows that many other biased and non-symmetric distributions also avoid the fixed point, and concludes that exponentially many inequivalent families of initializations exist. Numerical experiments suggest that first-moment-distinct initializations can lead to different attained minima.

Significance. If the first-moment diagnostic is shown to guarantee escape from exponential variance concentration, the result would clarify that initialization-based barren-plateau avoidance is highly non-unique and that the problem reduces to selecting among multiple trainable regions. The explicit recovery of known schemes and the operator-level comparison of biases are useful contributions. No machine-checked proofs or open reproducible code are referenced.

major comments (1)
  1. [Abstract] Abstract (central claim on exponentially many families): the first-moment diagnostic is presented as sufficient to escape the fully concentrated barren-plateau fixed point, yet barren plateaus are defined by exponential decay of Var[C] with qubit number; a non-zero first-moment bias does not by itself preclude the second moment from concentrating around the biased mean, and no derivation or numerical evidence is supplied showing that the full loss distribution escapes the concentrating regime.
minor comments (1)
  1. [Numerics] The numerical results are summarized without error bars, dataset sizes, or explicit description of the cost-function instances, which weakens the claim that different initializations reach distinct minima.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their careful reading and for identifying a key point of clarification needed in the abstract. We address the major comment below.

read point-by-point responses

  1. Referee: [Abstract] Abstract (central claim on exponentially many families): the first-moment diagnostic is presented as sufficient to escape the fully concentrated barren-plateau fixed point, yet barren plateaus are defined by exponential decay of Var[C] with qubit number; a non-zero first-moment bias does not by itself preclude the second moment from concentrating around the biased mean, and no derivation or numerical evidence is supplied showing that the full loss distribution escapes the concentrating regime.

    Authors: We agree that a non-zero first-moment bias alone does not guarantee that the variance of the loss will remain non-concentrated. The manuscript introduces the first-moment diagnostic specifically to detect escape from the fully concentrated fixed point (i.e., the operator-level bias is non-vanishing), and shows that this condition is satisfied by exponentially many inequivalent families, including but not limited to the known identity-initialization scheme that is already established to avoid barren plateaus. However, we do not claim or derive that every first-moment-nonzero distribution necessarily escapes exponential variance concentration; the abstract's phrasing that these initializations "avoid barren plateaus" is therefore imprecise. We will revise the abstract and the relevant discussion sections to state that the diagnostic identifies families that escape the zero-bias fully concentrated fixed point (a necessary condition), while noting that second-moment analysis would be required to confirm variance non-concentration in general. This is a clarification rather than a change to the core technical results on the first-moment operator comparison and the exponential count of families. revision: partial

Circularity Check

0 steps flagged

No circularity: first-moment diagnostic is an independent operator framework

full rationale

The paper introduces a new first-moment operator diagnostic to identify initializations that escape the concentrated barren-plateau fixed point, recovers known schemes such as identity and Gaussian initialization, and derives the existence of exponentially many inequivalent families directly from the framework's definitions and operator-level comparisons. No step reduces a claimed prediction or uniqueness result to a fitted parameter, self-referential definition, or load-bearing self-citation chain; the inequivalence and non-uniqueness claims are consequences of the diagnostic rather than tautological renamings or imported ansatzes. The derivation remains self-contained against external benchmarks for barren-plateau variance concentration.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The framework rests on standard quantum-information assumptions about ansatz structure and parameter distributions; no free parameters, invented entities, or ad-hoc axioms are visible from the abstract.

axioms (1)
  • domain assumption Concentration of the loss is governed by its first moment under the chosen initialization distribution.
    This is the load-bearing premise of the introduced diagnostic.

pith-pipeline@v0.9.1-grok · 5754 in / 1086 out tokens · 29300 ms · 2026-06-26T23:54:56.458584+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

113 extracted references · 26 canonical work pages · 4 internal anchors

  1. [1]

    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, Nature Reviews Physics3, 625–644 (2021)

    2021

  2. [2]

    Bharti, A

    K. Bharti, A. Cervera-Lierta, T. H. Kyaw, T. Haug, S. Alperin-Lea, A. Anand, M. Degroote, H. Heimonen, J. S. Kottmann, T. Menke,et al., Reviews of Modern Physics94, 015004 (2022)

    2022

  3. [3]

    S. Endo, Z. Cai, S. C. Benjamin, and X. Yuan, Journal of the Physical Society of Japan90, 032001 (2021)

    2021

  4. [4]

    Schuld, I

    M. Schuld, I. Sinayskiy, and F. Petruccione, Contempo- rary Physics56, 172 (2015)

    2015

  5. [5]

    Biamonte, P

    J. Biamonte, P. Wittek, N. Pancotti, P. Rebentrost, N. Wiebe, and S. Lloyd, Nature549, 195 (2017)

    2017

  6. [6]

    Challenges and opportu- nities in quantum machine learning,

    M. Cerezo, G. Verdon, H.-Y. Huang, L. Cincio, and P. J. Coles, Nature Computational Science 10.1038/s43588- 022-00311-3 (2022)

    work page doi:10.1038/s43588- 2022

  7. [7]

    Di Meglio, K

    A. Di Meglio, K. Jansen, I. Tavernelli, C. Alexandrou, S. Arunachalam, C. W. Bauer, K. Borras, S. Carrazza, A. Crippa, V. Croft,et al., Prx quantum5, 037001 (2024)

    2024

  8. [8]

    J. R. McClean, S. Boixo, V. N. Smelyanskiy, R. Babbush, and H. Neven, Nature Communications9, 1 (2018)

    2018

  9. [9]

    Larocca, S

    M. Larocca, S. Thanasilp, S. Wang, K. Sharma, J. Bia- monte, P. J. Coles, L. Cincio, J. R. McClean, Z. Holmes, and M. Cerezo, Nature Reviews Physics7, 174–189 (2025)

    2025

  10. [10]

    You and X

    X. You and X. Wu, inInternational Conference on Ma- chine Learning(PMLR, 2021) pp. 12144–12155

    2021

  11. [11]

    Fontana, M

    E. Fontana, M. Cerezo, A. Arrasmith, I. Rungger, and P. J. Coles, Quantum6, 804 (2022)

    2022

  12. [13]

    E. R. Anschuetz, International Conference on Learning Representations (2022)

    2022

  13. [14]

    Larocca, P

    M. Larocca, P. Czarnik, K. Sharma, G. Muraleedharan, P. J. Coles, and M. Cerezo, Quantum6, 824 (2022)

    2022

  14. [15]

    Bittel and M

    L. Bittel and M. Kliesch, Phys. Rev. Lett.127, 120502 (2021)

    2021

  15. [16]

    Cerezo, A

    M. Cerezo, A. Sone, T. Volkoff, L. Cincio, and P. J. Coles, Nature Communications12, 1 (2021)

    2021

  16. [17]

    Pesah, M

    A. Pesah, M. Cerezo, S. Wang, T. Volkoff, A. T. Sorn- borger, and P. J. Coles, Physical Review X11, 041011 (2021). 16

    2021

  17. [18]

    Khatri, R

    S. Khatri, R. LaRose, A. Poremba, L. Cincio, A. T. Sorn- borger, and P. J. Coles, Quantum3, 140 (2019)

    2019

  18. [19]

    Zhao and X.-S

    C. Zhao and X.-S. Gao, Quantum5, 466 (2021)

    2021

  19. [20]

    Liu, L.-W

    Z. Liu, L.-W. Yu, L.-M. Duan, and D.-L. Deng, Physical Review Letters129, 270501 (2022)

    2022

  20. [21]

    Miao and T

    Q. Miao and T. Barthel, Physical Review A109, L050402 (2024)

    2024

  21. [22]

    Bermejo, P

    P. Bermejo, P. Braccia, M. S. Rudolph, Z. Holmes, L. Cin- cio, and M. Cerezo, PRX Quantum7, 020304 (2026)

    2026

  22. [23]

    B. Bach, J. Falla, and I. Safro, in2024 IEEE Interna- tional Conference on Quantum Computing and Engineer- ing (QCE), Vol. 1 (IEEE, 2024) pp. 1–12

    2024

  23. [24]

    Larocca, F

    M. Larocca, F. Sauvage, F. M. Sbahi, G. Verdon, P. J. Coles, and M. Cerezo, PRX Quantum3, 030341 (2022)

    2022

  24. [25]

    J. J. Meyer, M. Mularski, E. Gil-Fuster, A. A. Mele, F. Arzani, A. Wilms, and J. Eisert, PRX Quantum4, 010328 (2023)

    2023

  25. [26]

    Skolik, M

    A. Skolik, M. Cattelan, S. Yarkoni, T. Bäck, and V. Dun- jko, npj Quantum Information9, 47 (2023)

    2023

  26. [27]

    Representation theory for geometric quantum machine learning (2022),

    M. Ragone, Q. T. Nguyen, L. Schatzki, P. Braccia, M. Larocca, F. Sauvage, P. J. Coles, and M. Cerezo, arXiv preprint arXiv:2210.07980 10.48550/arXiv.2210.07980 (2022)

    work page doi:10.48550/arxiv.2210.07980 2022

  27. [28]

    Q. T. Nguyen, L. Schatzki, P. Braccia, M. Ragone, P. J. Coles, F. Sauvage, M. Larocca, and M. Cerezo, PRX Quantum5, 020328 (2024)

    2024

  28. [29]

    Schatzki, M

    L. Schatzki, M. Larocca, Q. T. Nguyen, F. Sauvage, and M. Cerezo, npj Quantum Information10, 12 (2024)

    2024

  29. [30]

    Zheng, Z

    H. Zheng, Z. Li, J. Liu, S. Strelchuk, and R. Kondor, PRX Quantum4, 020327 (2023)

    2023

  30. [31]

    R. D. P. East, G. Alonso-Linaje, and C.-Y. Park, Quan- tum Science and Technology11, 025025 (2026)

    2026

  31. [32]

    S. Y. Chang, M. Larocca, and M. Cerezo, Manuscript in preparation (2026)

    2026

  32. [33]

    Tsvelikhovskiy, I

    B. Tsvelikhovskiy, I. Safro, and Y. Alexeev, IEEE Trans- actions on Quantum Engineering (2026)

    2026

  33. [34]

    R.Shaydulin, S.Hadfield, T.Hogg,andI.Safro,Quantum Information Processing20, 1 (2021)

    2021

  34. [35]

    Monbroussou, J

    L. Monbroussou, J. Landman, A. B. Grilo, R. Kukla, and E. Kashefi, Quantum9, 1745 (2025)

    2025

  35. [36]

    S. Raj, I. Kerenidis, A. Shekhar, B. Wood, J. Dee, S. Chakrabarti, R. Chen, D. Herman, S. Hu, P. Minssen, et al., Quantum7, 1191 (2023)

    2023

  36. [37]

    Fontana, D

    E. Fontana, D. Herman, S. Chakrabarti, N. Kumar, R. Yalovetzky, J. Heredge, S. H. Sureshbabu, and M. Pis- toia, Nature Communications15, 7171 (2024)

    2024

  37. [38]

    Ragone, B

    M. Ragone, B. N. Bakalov, F. Sauvage, A. F. Kemper, C. Ortiz Marrero, M. Larocca, and M. Cerezo, Nature Communications15, 7172 (2024)

    2024

  38. [39]

    N. L. Diaz, D. García-Martín, S. Kazi, M. Larocca, and M. Cerezo, arXiv preprint arXiv:2310.11505 10.48550/arXiv.2310.11505 (2023)

    work page doi:10.48550/arxiv.2310.11505 2023

  39. [40]

    M. T. West, J. Heredge, M. Sevior, and M. Usman, PRX Quantum5, 030320 (2024)

    2024

  40. [41]

    Heredge, N

    J. Heredge, N. Kumar, D. Herman, S. Chakrabarti, R. Yalovetzky, S. H. Sureshbabu, C. Li, and M. Pistoia, arXiv preprint arXiv:2405.08801 (2024)

    arXiv 2024

  41. [42]

    Aguilar, S

    G. Aguilar, S. Cichy, J. Eisert, and L. Bittel, arXiv preprint arXiv:2408.00081 10.48550/arXiv.2408.00081 (2024)

    work page doi:10.48550/arxiv.2408.00081 2024

  42. [43]

    Kökcü, R

    E. Kökcü, R. Wiersema, A. F. Kemper, and B. N. Bakalov, arXiv preprint arXiv:2409.19797 10.48550/arXiv.2409.19797 (2024)

    work page doi:10.48550/arxiv.2409.19797 2024

  43. [44]

    S. Kazi, M. Larocca, M. Farinati, P. J. Coles, M. Cerezo, and R. Zeier, PRX Quantum6, 040345 (2025)

    2025

  44. [45]

    Tsvelikhovskiy, B

    B. Tsvelikhovskiy, B. Bach, J. Falla, and I. Safro, arXiv preprint arXiv:2602.16141 (2026)

    Pith/arXiv arXiv 2026

  45. [46]

    Cerezo, M

    M. Cerezo, M. Larocca, D. García-Martín, N. L. Diaz, P. Braccia, E. Fontana, M. S. Rudolph, P. Bermejo, A. Ijaz, S. Thanasilp,et al., Nature Communications16, 7907 (2025)

    2025

  46. [47]

    Angrisani, A

    A. Angrisani, A. Schmidhuber, M. S. Rudolph, M. Cerezo, Z. Holmes, and H.-Y. Huang, Phys. Rev. Lett.135, 170602 (2025)

    2025

  47. [48]

    Lerch, R

    S. Lerch, R. Puig, M. Rudolph, A. Angrisani, T. Jones, M. Cerezo, S. Thanasilp, and Z. Holmes, arXiv preprint arXiv:2411.19896 10.48550/arXiv.2411.19896 (2024)

    work page doi:10.48550/arxiv.2411.19896 2024

  48. [49]

    E. R. Anschuetz and X. Gao, arXiv preprint arXiv:2402.08606 https://doi.org/10.48550/arXiv.2402.08606 (2024)

    work page doi:10.48550/arxiv.2402.08606 2024

  49. [50]

    A. A. Mele, A. Angrisani, S. Ghosh, S. Khatri, J. Eisert, D. Stilck França, and Y. Quek, Nature Physics , 1 (2026)

    2026

  50. [51]

    S. Shin, Y. S. Teo, and H. Jeong, Phys. Rev. Res.6, 023218 (2024)

    2024

  51. [52]

    Ermakov, O

    I. Ermakov, O. Lychkovskiy, and T. Byrnes, arXiv preprint arXiv:2401.08187 10.48550/arXiv.2401.08187 (2024)

    work page doi:10.48550/arxiv.2401.08187 2024

  52. [53]

    Simulation of

    A. Miller, J. Favre, Z. Holmes, Ö. Salehi, R. Chakraborty, A. Nykänen, Z. Zimboras, A. Glos, and G. García-Pérez, arXiv preprint arXiv:2503.18939 10.48550/arXiv.2503.18939 (2025)

    work page doi:10.48550/arxiv.2503.18939 2025

  53. [54]

    M. S. Rudolph, A. Angrisani, A. Wright, I. Sanderski, R. Puig, and Z. Holmes, arXiv preprint arXiv:2602.04878 https://doi.org/10.48550/arXiv.2602.04878 (2026)

    work page doi:10.48550/arxiv.2602.04878 2026

  54. [55]

    Enabling Lie-Algebraic Classical Simulation beyond Free Fermions

    A. Bärligea, M. L. Sims-Goh, and J. S. Kottmann, arXiv preprint arXiv:2604.16701 10.48550/arXiv.2604.16701 (2026)

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2604.16701 2026

  55. [56]

    Grant, L

    E. Grant, L. Wossnig, M. Ostaszewski, and M. Benedetti, Quantum3, 214 (2019)

    2019

  56. [57]

    Kulshrestha and I

    A. Kulshrestha and I. Safro, in2022 IEEE Interna- tional Conference on Quantum Computing and Engineer- ing (QCE)(IEEE, 2022) pp. 197–203

    2022

  57. [58]

    Duffield, M

    S. Duffield, M. Benedetti, and M. Rosenkranz, Machine Learning: Science and Technology4, 025007 (2023)

    2023

  58. [59]

    Heyraud, Z

    V. Heyraud, Z. Li, K. Donatella, A. L. Boité, and C. Ciuti, PRX Quantum4, 040335 (2023)

    2023

  59. [60]

    Zhang, L

    K. Zhang, L. Liu, M.-H. Hsieh, and D. Tao, inAdvances in Neural Information Processing Systems(2022)

    2022

  60. [61]

    Park and N

    C.-Y. Park and N. Killoran, Quantum8, 1239 (2024)

    2024

  61. [62]

    Y. Wang, B. Qi, C. Ferrie, and D. Dong, Physical Review Applied22, 054005 (2024)

    2024

  62. [63]

    C.-Y. Park, M. Kang, and J. Huh, arXiv preprint arXiv:2403.04844 (2024). 17

    arXiv 2024

  63. [64]

    Shi and Y

    X. Shi and Y. Shang, arXiv preprint arXiv:2402.13501 (2024)

    arXiv 2024

  64. [65]

    R. Puig, M. Drudis, S. Thanasilp, and Z. Holmes, PRX Quantum6, 010317 (2025)

    2025

  65. [66]

    R. Puig, B. Casas, A. Cervera-Lierta, Z. Holmes, and A. Pérez-Salinas, arXiv preprint arXiv:2602.06137 https://doi.org/10.48550/arXiv.2602.06137 (2026)

    work page doi:10.48550/arxiv.2602.06137 2026

  66. [67]

    IQP Born machines under data-dependent and agnostic initialization strategies,

    S. Lerch, J. Bowles, R. Puig, E. Armengol, Z. Holmes, and S. Thanasilp, arXiv preprint arXiv:2603.14576 10.48550/arXiv.2603.14576 (2026)

    work page doi:10.48550/arxiv.2603.14576 2026

  67. [68]

    E. R. Anschuetz and B. T. Kiani, Nature Communications 13, 7760 (2022)

    2022

  68. [69]

    N. A. Nemkov, E. O. Kiktenko, and A. K. Fedorov, arXiv preprint arXiv:2405.05332 10.48550/arXiv.2405.05332 (2024)

    work page doi:10.48550/arxiv.2405.05332 2024

  69. [70]

    Kandala, A

    A. Kandala, A. Mezzacapo, K. Temme, M. Takita, M. Brink, J. M. Chow, and J. M. Gambetta, Nature549, 242 (2017)

    2017

  70. [71]

    H. R. Grimsley, S. E. Economou, E. Barnes, and N. J. Mayhall, Nature Communications10, 1 (2019)

    2019

  71. [72]

    A Quantum Approximate Optimization Algorithm

    E. Farhi, J. Goldstone, and S. Gut- mann, arXiv preprint arXiv:1411.4028 https://doi.org/10.48550/arXiv.1411.4028 (2014)

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1411.4028 2014

  72. [73]

    Wiersema, C

    R. Wiersema, C. Zhou, Y. de Sereville, J. F. Carrasquilla, Y. B. Kim, and H. Yuen, PRX Quantum1, 020319 (2020)

    2020

  73. [74]

    S. Y. Chang, M. Grossi, B. Le Saux, and S. Vallecorsa, in 2023 IEEE International Conference on Quantum Com- puting and Engineering (QCE), Vol. 01 (2023) pp. 229– 235

    2023

  74. [75]

    R. T. Forestano, M. C. Cara, G. R. Dahale, Z. Dong, S. Gleyzer, D. Justice, K. Kong, T. Magorsch, K. T. Matchev, K. Matcheva,et al., arXiv preprint arXiv:2311.18672 (2023)

    arXiv 2023

  75. [76]

    M. T. West, M. Sevior, and M. Usman, Machine Learning: Science and Technology4, 035027 (2023)

    2023

  76. [77]

    H.Zheng, Z.Li, J.Liu, S.Strelchuk,andR.Kondor,arXiv preprint arXiv:2207.07250 (2022)

    arXiv 2022

  77. [78]

    Mhiri, R

    H. Mhiri, R. Puig, S. Lerch, M. S. Rudolph, T. Chotibut, S. Thanasilp, and Z. Holmes, arXiv preprint arXiv:2502.07889 https://doi.org/10.48550/arXiv.2502.07889 (2025)

    work page doi:10.48550/arxiv.2502.07889 2025

  78. [79]

    Peruzzo, J

    A. Peruzzo, J. McClean, P. Shadbolt, M.-H. Yung, X.-Q. Zhou, P. J. Love, A. Aspuru-Guzik, and J. L. O’brien, Nature Communications5, 1 (2014)

    2014

  79. [80]

    S. Sim, P. D. Johnson, and A. Aspuru-Guzik, Advanced Quantum Technologies2, 1900070 (2019)

    2019

  80. [81]

    Holmes, K

    Z. Holmes, K. Sharma, M. Cerezo, and P. J. Coles, PRX Quantum3, 010313 (2022)

    2022

Showing first 80 references.