Pith. sign in

REVIEW 5 major objections 4 minor 1 cited by

Continuous-time quantum walk-based ans\"atze on neutral atom hardware

T0 review · 5 major / 4 minor · reviewed 2026-08-05 · deepseek-v4-flash

Pith's one-line read Quantum walks on neutral atoms reproduce the amplification signatures of efficient quantum-walk protocols at low depth.

desk verdict A genuine first implementation of CTQW ansätze on analog neutral-atom hardware, with a solid core but an abstract that overstates the hardware verification of the bracelet-state scaling. read the letter →

arxiv 2509.00386 v3 pith:JRZZTKSD submitted 2025-08-30 quant-ph

classification quant-ph MSC 68Q1281P68 PACS 03.67.Lx
keywords continuous-timequantumwalkRydbergblockadeneutralatomarrayphase-walkansatzLucascubebraceletstatesamplificationscalingstatepreparation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tries to show that continuous-time quantum walks (CTQWs) on constrained independent-set graphs can be turned into practical state-preparation ansätze on analog neutral-atom hardware. It maps the abstract walk generator onto the Rydberg Hamiltonian, using the Rydberg blockade as a physical projector onto the subspace of valid bitstrings. For product-state targets it derives closed-form near-optimal walk times, and for 'bracelet' targets—equal-weight superpositions over dihedrally symmetric bitstring orbits—it introduces an optimization protocol whose required evolution time scales inversely with the spectral gap rather than its square. On current hardware the amplified success probabilities follow power laws whose effective polynomial order exceeds two at low depth, the signature of an efficient quantum-walk protocol. The authors conclude that the mechanisms behind CTQW speedups are already observable on noisy analog processors, even though the experiments do not by themselves establish an algorithmic speedup.

What carries the argument

The central object is the phase-walk ansatz |ψ⟩ = ∏_q e^{−iτ_q G} e^{−iγ_q C}|ψ0⟩, with G the adjacency walk on the Lucas cube (the graph of independent sets of a ring) and C a diagonal phasor. The paper realizes G through the Rydberg Hamiltonian: the blockade acts as a projector P, so G = Σ_i P σ_x^i P, and global or local phase jumps implement C. Two analytic structures carry the argument: (1) restriction of the walk to the positive subspace of a target bitstring yields an effective SU(2) spin-chain ladder with couplings J_{j,j+1}=√((k−j)(j+1)), which predicts near-perfect transfer with optimized times; (2) a frequency-resolution model on the dihedrally invariant subspace predicts bracelet

What would settle it

Prepare a product state at depth p=3 on the same device while measuring and correcting the per-site local-detuning phase error to below 1%; if the effective polynomial order n does not rise back above the p=2 value, the claimed super-quadratic convergence is not robust under error reduction. Alternatively, compare the emulated Rydberg evolution against the ideal CTQW for walk times beyond those used in the paper and look for fidelity breakdown from blockade violations.

Watch

Extended reading notes

Core claim

The central discovery, on the paper's own terms, is that a phase-walk ansatz—alternating a constrained quantum-walk mixer with a diagonal phasor—can prepare both unentangled product states and entangled bracelet states in the Rydberg-blockaded subspace, and does so with scaling that matches ideal CTQW predictions. For product states, the walk generator restricted to the positive subspace of the target bitstring reduces to an effective SU(2) spin chain, so the target is reached by constructive interference with an effective coupling Jeff; this yields closed-form analytic starting points for variational optimization. For bracelet states, the relevant timescale is set by the smallest resolvable

Load-bearing premise

The hardware experiment reproduces the ideal CTQW only if the Rydberg blockade acts as a near-perfect projector onto the independent-set subspace; residual blockade violations and van der Waals tails outside the unit disk create an error Hamiltonian that can break the mapping at longer evolution times.

Editorial extensions

If this is right

  • If the CTQW-to-Rydberg mapping is faithful, any independent-set walk generator on a unit-disk graph can be programmed by placing atoms at blockade distances, making the blockade a reusable resource for constrained-subspace evolution.
  • The closed-form product-state parameters transfer to hardware with minimal calibration, so scaling tests and benchmarking can be done without expensive optimization loops.
  • Because bracelet-state time scales as τeff ∝ 1/Δmin instead of 1/Δmin², CTQW-based preparation can outpace adiabatic protocols for the same target on the same device.
  • The super-quadratic effective order n observed at low depth means that the constructive-interference mechanism, not just graph size, drives success probability; as hardware errors drop, theory predicts n to rise toward ideal CTQW values.
  • Quench dynamics that distinguish coherent from incoherent bracelet states give a practical fidelity witness for entangled-state preparation that does not require full tomography.

Reading between the lines

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

  • Beyond the paper: if per-site local-detuning phase errors are suppressed, the depth-p=3 hardware exponent for product states should recover and exceed the p=2 value; this is a direct, testable prediction of the paper's error model.
  • Beyond the paper: the same spectral-gap control protocol could prepare scarred or symmetry-protected resource states on other constraint graphs, where the ratio of nearest to next-nearest distances is less favourable and the blockade-radius optimization becomes the key bottleneck.
  • Beyond the paper: a natural extension is to use these ansätze for hybrid optimization where the target is not known in advance; the fast-forward product-state protocol suggests a warm-start strategy based on closed-form effective couplings.
  • Beyond the paper: the τeff ∝ 1/Δmin law, if it persists at larger N, implies that only a small resolvable spectral gap of the Lucas cube needs to be engineered, guiding future Hamiltonian-engineering approaches to tailor walk graphs.
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

5 major / 4 minor

Summary. The paper reports an implementation of continuous-time quantum walk (CTQW) based variational ansätze on QuEra's Aquila analog neutral-atom processor. It targets product states and 'bracelet' symmetric states in the independent-set subspace of a ring graph. For product states, the authors derive analytic starting parameters for a phase-walk ansatz using a second-order Baker-Campbell-Hausdorff argument, transfer them directly to the Rydberg Hamiltonian, and report super-quadratic amplification A=c|V|^α on hardware. For bracelet states, they propose a spectral-gap frequency-resolution model predicting τ_eff ∝ 1/Δ_min, and claim verification on Aquila. The paper includes noiseless Rydberg emulation, Aquila hardware data with Bayesian readout-error mitigation, and quench-dynamics evidence for coherence.

Significance. If the central claims are correct, the paper would be a valuable demonstration that CTQW-based ansätze can be mapped onto current analog neutral-atom hardware, that closed-form parameter choices can be transferred with minimal calibration, and that characteristic amplification mechanisms of efficient quantum walks are visible on noisy hardware. The study's strengths include the direct parameter transfer from ideal CTQW to Rydberg dynamics, the comparison across perfect, emulated, and hardware regimes, the use of quench dynamics to probe coherence, and the careful handling of measurement errors. However, the quantitative speedup claims rest on fits with free exponents and thresholds calibrated on the same data they are used to explain, and the abstract's hardware-verification claim for the bracelet-state scaling is not supported by the data shown. These issues need to be addressed before the stronger conclusions can be accepted.

major comments (5)
  1. [Abstract and §IV.B, Fig. 9] The abstract states that the brace-state inverse-gap scaling τ_eff ∝ 1/Δ_min is 'verified on Aquila.' The hardware data in Fig. 9 contain no τ_eff vs 1/Δ_min comparison; the inverse-gap relation is shown only for ideal CTQW dynamics in Fig. 3 and in the CTQW/Rydberg emulation ratio in Fig. 8. Section IV.B explicitly concedes that for [z_half] the hardware results 'do not reproduce this trend,' and for [z_MIS] only 'qualitatively consistent' amplification is claimed, adding that alternative parameter sets at shorter times cannot be precluded. The abstract overstates the experimental support; the claims need to be revised to distinguish theoretical/emulated verification from hardware consistency.
  2. [§II.D, §IV.A, Eqs. (13)-(14)] The super-quadratic convergence claim for product states is quantified by fitting A=c|V|^α to the measured success probabilities and then converting to n=1/(1-α). Because α and c are fitted to the same data used to report the speedup order, the 'prediction' of the speedup order restates the fitted exponent. The hardware p=3 result (n=3.76, down from n=8.07 at p=2) shows that the fitted order is not stable. To support a predictive claim, the exponent should be derived from the analytic walk parameters or fixed before the hardware runs, and then tested on the data; otherwise the fit should be described as a descriptive summary rather than evidence of super-quadratic convergence.
  3. [§II.E.1, Eqs. (21)-(31)] The derivation of τ_0* and τ_1* uses a second-order BCH truncation and introduces graph-dependent constants κ_leak and κ_ret without giving closed forms (Eqs. (29)-(31)). The paper calls these 'closed-form expressions,' but κ=κ_leak/κ_ret is not specified analytically or tabulated, so the analytic starting point cannot be reproduced from the text. Please provide the formulas or numerical values for κ_leak and κ_ret, or soften the closed-form claim.
  4. [§II.E.2, Eqs. (36)-(37), Fig. 3] The definition Δ_min(κ)=min_{λ_rs≥κ/τ_eff} λ_rs makes Δ_min a function of τ_eff itself. For any spectrum with resolvable eigenvalues near the threshold, Δ_min ≈ κ/τ_eff, so Eq. (37), τ_eff ∝ 1/Δ_min, becomes close to an identity rather than a physical prediction. Additionally, κ* ≈ 7.2 is calibrated by optimizing mean(r)-2std(r) on the same simulation data (Fig. 3 caption) that is then used to demonstrate the linear scaling. This circularity needs to be removed by defining Δ_min from the bare spectrum independently of the measured τ_eff and by validating the model on data not used to choose κ*.
  5. [§III, Appendix B, Fig. 12] The hardware implementation relies on treating the Rydberg blockade as a near-perfect projector onto the independent-set subspace. Appendix B gives the error Hamiltonian H_err, and Fig. 12 shows emulated fidelities below unity. Section IV.B notes that van der Waals interactions accumulate phase errors over the long τ_eff used for bracelet states, with hardware success falling roughly linearly in τ_eff. This error budget is not propagated into the fitted amplification exponents or the inverse-gap scaling claim. Please provide quantitative estimates of how H_err affects A and τ_eff, or explicitly restrict the hardware claims to qualitative consistency.
minor comments (4)
  1. [Fig. 3 caption] The caption contains an incomplete sentence: 'Here ∆ min(κ).' Please complete or remove it.
  2. [§V, Eq. (45)] The text says there are 'nontrivial off-diagonal matrix elements in ρinc and (if prepared exactly) none in ρinc.' The second occurrence should be 'ρcoh'; as written it is a typo that obscures the argument.
  3. [Table III header] Table III is labeled 'product states' in the caption, but it reports bracelet states. The label should be corrected.
  4. [§III, Appendix B] The reference to 'the supplemental of [50]' is vague; please give a specific appendix or equation number, or include the relevant derivation.

Circularity Check

2 steps flagged · score 6.0 of 10

Bracelet inverse-gap scaling is definitional in Δ_min(κ), and the product-state speedup order is a transform of the fitted α; the claimed hardware verification is overstated.

  1. self definitional [Section II.E.2, Eqs. (36)-(37); Fig. 3 caption]
    "We encode this threshold by introducing the dimensionless constant κ = Θ(1) and enforcing λrsτeff ≥ κ. Among resolvable pairs, we then define the resolvable spectral minimum, Δmin(κ) = min_{λrs≥κ/τeff} λrs, and the model predicts that this low-frequency component sets the timescale, τeff ∝ 1/Δmin(κ)."

    Because Δmin(κ) is defined as the smallest gap satisfying λrs ≥ κ/τeff, the definition itself enforces Δmin(κ) ≥ κ/τeff; if any resolvable gap sits near this threshold, τeff ≈ κ/Δmin follows automatically. The linear plot of τeff vs 1/Δmin is therefore a consequence of how Δmin was defined, not an independent discovery of the scaling law. The constant κ* ≈ 7.2 is then calibrated on the same simulation data ('maximizing mean(r)−2 std(r) over sliding κ windows and taking the window center'), so the agreement in Fig. 3 is a fit of κ. The subsequent hardware claim is also unsupported: Fig. 9 plots success vs τeff and amplification vs |V|, not τeff vs 1/Δmin, and Section IV.B concedes that for [zhalf] 'the hardware results do not reproduce this trend.'

  2. fitted input called prediction [Section IV, fitting paragraph; Eqs. (13)-(14)]
    "In Section IV A, we fit the amplification A to Eq. (13) via weighted nonlinear least squares, taking the 95% CI for α from the fit covariance under the constraint α <1. The intervals are propagated through n = 1/(1 − α), with n → ∞ (reported as “ ≥”) when the upper bound reaches one."

    The speedup order n is not an independent quantity: Eq. (14) defines n = 1/(1−α), so n is a one-to-one transform of the fitted exponent α. The paper fits α to the success probabilities via Eq. (13) and then quotes n = {8.17, 15.3, ...} as 'effective polynomial speedup orders'; the super-quadratic convergence claim (n > 2) is exactly the fitted inequality α > 1/2. No analytic calculation in the paper derives α for these product-state ansätze; the τ0*, τ1* derivation fixes angles, not the scaling exponent. Hence the headline convergence order restates the fitted power-law slope rather than confirming a pre-existing prediction.

full rationale

The paper contains substantial independent content: the analytic derivation of near-optimal τ0* and τ1* for product states, the Rydberg-blockade mapping with its perturbative error analysis, the noiseless emulation, and the quench-based coherence check are all self-contained and not circular. The self-citations (e.g., [8] for the A = c|V|^α metric) are not load-bearing uniqueness claims; they merely introduce a standard power-law fitting convention. However, two central 'predictions' reduce to their own inputs. First, the bracelet-state relation τeff ∝ 1/Δ_min(κ) is built into the definition of Δ_min(κ), which is the minimum gap above κ/τeff, and the constant κ* is calibrated on the same dataset used to demonstrate the linear scaling. Second, the product-state 'effective polynomial speedup order' n is defined as 1/(1−α), so the observed super-quadratic convergence is a relabeling of the fitted exponent α, not a separately derived prediction. There is also an explicit overstatement: the abstract says the inverse-gap scaling is 'verified on Aquila,' but Section IV.B and Fig. 9 show hardware success decreasing with τeff, and the paper admits 'we cannot preclude the existence of alternative parameter sets that achieve equivalent or superior performance at shorter effective walk times.' That overstatement is a correctness risk rather than a circular step, but it further weakens the central scaling claim. Overall, the partial circularity is real and concentrated in the scaling claims, while the experimental implementation and coherence verification remain independently valuable, so a score of 6 is appropriate.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The paper introduces no new particles or forces. It introduces a heuristic spectral model with a fitted threshold κ* and uses fitted amplification exponents α to quantify speedup. The physical mapping and measurement model are standard domain assumptions.

free parameters (2)
  • kappa* (spectral-gap model threshold) = ~7.2 ± 0.4
    Introduced in Section II.E.2 to define the resolvable spectral minimum Δ_min(κ) and enforce λ_rs τ_eff ≥ κ. Calibrated by maximizing mean(r)-2 std(r) over sliding κ windows in the data of Fig 3 bottom, then used to demonstrate τ_eff ∝ 1/Δ_min.
  • Amplification exponent α and prefactor c = e.g., hardware zhalf p=1: c=0.595, α=0.729±0.041; per scenario and depth
    Fitted via weighted nonlinear least squares to A = c|V|^α in Section IV.A; the effective speedup order n = 1/(1-α) is computed from this fitted exponent, so the super-quadratic claim restates a fitted parameter.
assumptions (3)
  • domain assumption Rydberg blockade projects evolution onto the independent-set subspace with negligible perturbative corrections
    Section III and Appendix B: the hardware maps the walk generator G to H(t) assuming doubly-excited states are suppressed (blockade radius rb = η√(rmin rmax)). If Herr (Eq. B2) is not small, the physical dynamics deviate from the CTQW.
  • domain assumption Measurement errors are described by an independent asymmetric bit-flip channel with known parameters (P00,P11) = (0.99,0.93) or (0.90,0.93)
    Appendix A: EM reconstruction relies on this model to correct raw counts; wrong channel parameters would bias all reported success probabilities.
  • ad hoc to paper The frequency-resolution model with κ = Θ(1) sets the required time as τ_eff ∝ 1/Δ_min(κ)
    Section II.E.2: a heuristic, not proven, scaling law; κ is calibrated to the simulation data, so the claimed advantage over adiabatic protocols depends on this model being correct.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Continuous-time quantum walk-based ans\"atze on neutral atom hardware." pith.science (2026). https://pith.science/paper/JRZZTKSD

@misc{pith2026250900386,
  author       = {Pith},
  title        = {Pith review of: Continuous-time quantum walk-based ans\"atze on neutral atom hardware},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JRZZTKSD}},
  note         = {Machine review of arXiv:2509.00386}
}
read the original abstract

Continuous-time quantum walks offer provable speedups for certain computational problems, yet translating these advantages to near-term hardware remains challenging. We realize variational ans\"atze based on continuous-time quantum walks on an analog neutral-atom processor. For unentangled targets, we derive closed-form expressions for near-optimal control parameters that transfer directly to hardware with minimal calibration. On QuEra's Aquila processor we observe the super-quadratic convergence characteristic of efficient quantum walk algorithms, visible at low circuit depth, with theory predicting stronger speedups as hardware improves. For entangled targets, specifically symmetric superpositions in the Rydberg-blockaded subspace, we introduce an optimization protocol exploiting spectral properties of the walk dynamics. The required evolution time scales inversely with the spectral gap, offering an advantage over adiabatic protocols, whose evolution time scales as the inverse square of the spectral gap. We verify this scaling behavior on Aquila and confirm that the prepared states are coherent superpositions via quench dynamics. Our results establish a practical pathway from abstract quantum walk algorithms to analog quantum processors, demonstrating that the dynamics underlying their potential for super-quadratic quantum speedup are accessible on current devices.

Figures

Figures reproduced from arXiv: 2509.00386 by the authors.

Figure 1
Figure 1. FIG. 1. An example walk graph for independent set con [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. Success probability of preparing product states [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Bracelet state preparation. Top: Population of the [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: FIG. 4. Representative Analog programs to implement quan [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5. Probability distributions for preparation of the [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 6
Figure 6. Figure 6: FIG. 6. Comparison of success probabilities [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]
Figure 7
Figure 7. Figure 7: FIG. 7. Ratio of success probabilities for preparation of [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]
Figure 8
Figure 8. Figure 8: FIG. 8. Comparison between the ideal CTQW-based [PITH_FULL_IMAGE:figures/full_fig_p012_8.png]
Figure 9
Figure 9. Figure 9: FIG. 9. Performance for bracelet state preparation of [PITH_FULL_IMAGE:figures/full_fig_p013_9.png]
Figure 10
Figure 10. Figure 10: FIG. 10. Quenches for target state [PITH_FULL_IMAGE:figures/full_fig_p014_10.png]
Figure 11
Figure 11. Figure 11: FIG. 11. An example unit disk graph to illustrate [PITH_FULL_IMAGE:figures/full_fig_p018_11.png]
Figure 12
Figure 12. Figure 12: FIG. 12. Emulated fidelity for approximation of a CTQW [PITH_FULL_IMAGE:figures/full_fig_p019_12.png]

Discussion (0). Sign in 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. Neutral Atom Quantum Computing: Principles, Routes, Progress, and Challenges

    quant-ph 2026-08 unverdicted novelty 2.0 of 10

    A broad review of neutral atom quantum computing covering principles, technical routes, 2000-2026 achievements, industry status, and bottlenecks.

Reference graph

Works this paper leans on

65 extracted references · 56 canonical work pages · cited by 1 Pith paper

  1. [1]

    Preskill, Quantum 2, 79 (2018)

    J. Preskill, Quantum 2, 79 (2018)

  2. [2]

    Perfect”), noiseless emulation of the Rydberg Hamiltonian (“Emulation

    Bracelet states The optimization strategy for preparation of bracelet states is guided by two practical considerations, as ex- panded in Section III. First, we aim to minimize the total evolution time and, second, utilize the hardware-native realization of the phasor unitary ˆUh(γ) = e−iγ ˆn, (33) where ˆn = PN i=1 |1⟩i⟨1|i applies a uniform phase to all ...

  3. [3]

    (31) and (32) is indicated by a yellow ‘+’

    The optimal points (given in Table II) are marked with a red ‘X’, and theoretical predictions from Eqs. (31) and (32) is indicated by a yellow ‘+’

  4. [4]

    Farhi, J

    E. Farhi, J. Goldstone, and S. Gutmann, A quan- tum approximate optimization algorithm (2014), arXiv:1411.4028 [quant-ph]

  5. [5]

    Ambainis, SIAM Journal on Computing 37, 210 (2007)

    A. Ambainis, SIAM Journal on Computing 37, 210 (2007)

  6. [6]

    A. M. Childs, Physical Review Letters 102, 180501 (2009)

  7. [7]

    A. M. Childs, R. Cleve, E. Deotto, E. Farhi, S. Gutmann, and D. A. Spielman, in Proceedings of the thirty-fifth an- nual ACM symposium on Theory of computing , STOC ’03 (Association for Computing Machinery, San Diego, CA, USA, 2003) pp. 59–68

  8. [8]

    Marsh and J

    S. Marsh and J. B. Wang, Physical Review Research 2, 023302 (2020)

Show all 65 references
  1. [9]

    Hadfield, Z

    S. Hadfield, Z. Wang, B. O’gorman, E. G. Rieffel, D. Ven- turelli, and R. Biswas, Algorithms 12, 34 (2019)

  2. [10]

    Matwiejew, J

    E. Matwiejew, J. Pye, and J. B. Wang, Quantum Science and Technology 8, 045013 (2023)

  3. [11]

    Bennett, L

    T. Bennett, L. Noakes, and J. Wang, in 2024 IEEE In- 15 ternational Conference on Quantum Computing and En- gineering (QCE) (IEEE, 2024) p. 31–41

  4. [12]

    Deffner and S

    S. Deffner and S. Campbell, Journal of Physics A: Math- ematical and Theoretical 50, 453001 (2017)

  5. [13]

    Christandl, N

    M. Christandl, N. Datta, A. Ekert, and A. J. Landahl, Physical Review Letters 92, 187902 (2004)

  6. [14]

    A. M. Childs and J. Goldstone, Physical Review A 70, 022314 (2004)

  7. [15]

    Gu´ ery-Odelin, A

    D. Gu´ ery-Odelin, A. Ruschhaupt, A. Kiely, E. Tor- rontegui, S. Mart ´ ınez-Garaot, and J. Muga, Reviews of Modern Physics 91, 045001 (2019)

  8. [16]

    Peruzzo, M

    A. Peruzzo, M. Lobino, J. C. F. Matthews, N. Matsuda, A. Politi, K. Poulios, X.-Q. Zhou, Y. Lahini, N. Ismail, K. W¨ orhoff, Y. Bromberg, Y. Silberberg, M. G. Thomp- son, and J. L. OBrien, Science 329, 1500–1503 (2010)

  9. [17]

    P. M. Preiss, R. Ma, M. E. Tai, A. Lukin, M. Rispoli, P. Zupancic, Y. Lahini, R. Islam, and M. Greiner, Science 347, 1229–1233 (2015)

  10. [18]

    Tamura, T

    M. Tamura, T. Mukaiyama, and K. Toyoda, Physical Re- view Letters 124, 200501 (2020)

  11. [19]

    Yan, Y.-R

    Z. Yan, Y.-R. Zhang, M. Gong, Y. Wu, Y. Zheng, S. Li, C. Wang, F. Liang, J. Lin, Y. Xu, C. Guo, L. Sun, C.-Z. Peng, K. Xia, H. Deng, H. Rong, J. Q. You, F. Nori, H. Fan, X. Zhu, and J.-W. Pan, Science 364, 753–756 (2019)

  12. [20]

    M. Gong, S. Wang, C. Zha, M.-C. Chen, H.-L. Huang, Y. Wu, Q. Zhu, Y. Zhao, S. Li, S. Guo, H. Qian, Y. Ye, F. Chen, C. Ying, J. Yu, D. Fan, D. Wu, H. Su, H. Deng, H. Rong, and et al., Science 372, 948–952 (2021)

  13. [21]

    A. W. Young, W. J. Eckner, N. Schine, A. M. Childs, and A. M. Kaufman, Science 377, 885 (2022)

  14. [22]

    D. Qu, E. Matwiejew, K. Wang, J. Wang, and P. Xue, Quantum Science and Technology 9, 025014 (2024)

  15. [23]

    S. Choi, C. J. Turner, H. Pichler, W. W. Ho, A. A. Michailidis, Z. Papi´ c, M. Serbyn, M. D. Lukin, and D. A. Abanin, Physical Review Letters 122, 220603 (2019)

  16. [24]

    Bernien, S

    H. Bernien, S. Schwartz, A. Keesling, H. Levine, A. Om- ran, H. Pichler, S. Choi, A. S. Zibrov, M. Endres, M. Greiner, V. Vuleti´ c, and M. D. Lukin, Nature 551, 579–584 (2017)

  17. [25]

    Wurtz, A

    J. Wurtz, A. Bylinskii, B. Braverman, J. Amato-Grill, S. H. Cantu, F. Huber, A. Lukin, F. Liu, P. Wein- berg, J. Long, S.-T. Wang, N. Gemelke, and A. Keesling, Aquila: Quera’s 256-qubit neutral-atom quantum com- puter (2023)

  18. [26]

    S. E. Venegas-Andraca, Quantum Information Process- ing 11, 1015 (2012)

  19. [27]

    Hadfield, Z

    S. Hadfield, Z. Wang, B. O’Gorman, E. Rieffel, D. Ven- turelli, and R. Biswas, Algorithms 12, 34 (2019)

  20. [28]

    Munarini, C

    E. Munarini, C. P. Cippo, and N. Z. Salvi, The Fibonacci Quarterly 39, 12 (2001)

  21. [29]

    L. G. Valiant, SIAM Journal on Computing 8, 410–421 (1979)

  22. [30]

    A. R. Ashrafi, J. Azarija, K. Fathalikhani, S. Klav zar, and M. Petkovsek, Annals of Combinatorics 20, 209–229 (2016)

  23. [31]

    Gonzales, R

    A. Gonzales, R. Herrman, C. Campbell, I. Gaidai, J. Liu, T. Tomesh, and Z. H. Saleem, npj Quantum Information 11, 10.1038/s41534-025-01093-y (2025)

  24. [32]

    Sawada, SIAM Journal on Computing 31, 259–268 (2001)

    J. Sawada, SIAM Journal on Computing 31, 259–268 (2001)

  25. [33]

    Wurtz and P

    J. Wurtz and P. Love, Physical Review A 103, 042612 (2021)

  26. [34]

    Brassard, P

    G. Brassard, P. Høyer, M. Mosca, and A. Tapp, Quantum Computation and Information , 53–74 (2002)

  27. [35]

    Ambainis, International Journal of Quantum Informa- tion 01, 507–518 (2003)

    A. Ambainis, International Journal of Quantum Informa- tion 01, 507–518 (2003)

  28. [36]

    Zalka, Physical Review A 60, 2746–2751 (1999)

    C. Zalka, Physical Review A 60, 2746–2751 (1999)

  29. [37]

    Wurtz and P

    J. Wurtz and P. J. Love, Quantum 6, 635 (2022)

  30. [38]

    Krovi and T

    H. Krovi and T. A. Brun, Physical Review A 75, 062332 (2007)

  31. [39]

    Atia and S

    Y. Atia and S. Chakraborty, Phys. Rev. A 104, 032215 (2021)

  32. [40]

    Desaules, K

    J.-Y. Desaules, K. Bull, A. Daniel, and Z. Papi´ c, Physical Review B 105, 245137 (2022)

  33. [41]

    Saffman, Journal of Physics B: Atomic, Molecular and Optical Physics 49, 202001 (2016)

    M. Saffman, Journal of Physics B: Atomic, Molecular and Optical Physics 49, 202001 (2016)

  34. [42]

    Morgado and S

    M. Morgado and S. Whitlock, A VS Quantum Science 3, 023501 (2021)

  35. [43]

    Bluvstein, S

    D. Bluvstein, S. J. Evered, A. A. Geim, S. H. Li, H. Zhou, T. Manovitz, S. Ebadi, M. Cain, M. Kalinowski, D. Hangleiter, J. P. Bonilla Ataides, N. Maskara, I. Cong, X. Gao, P. Sales Rodriguez, T. Karolyshyn, G. Semegh- ini, M. J. Gullans, M. Greiner, V. Vuleti´ c, and M. D. Lu...

  36. [44]

    B. W. Reichardt, A. Paetznick, D. Aasen, I. Basov, J. M. Bello-Rivas, P. Bonderson, R. Chao, W. van Dam, M. B. Hastings, R. V. Mishmash, A. Paz, M. P. da Silva, A. Sundaram, K. M. Svore, A. Vaschillo, Z. Wang, M. Zanner, W. B. Cairncross, C.-A. Chen, D. Crow, and et al., Fault...

  37. [45]

    P. S. Rodriguez, J. M. Robinson, P. N. Jepsen, Z. He, C. Duckering, C. Zhao, K.-H. Wu, J. Campo, K. Bag- nall, M. Kwon, T. Karolyshyn, P. Weinberg, M. Cain, S. J. Evered, A. A. Geim, M. Kalinowski, S. H. Li, T. Manovitz, J. Amato-Grill, J. I. Basham, and et al., Nature 10.1038...

  38. [46]

    Bluvstein, A

    D. Bluvstein, A. A. Geim, S. H. Li, S. J. Evered, J. P. B. Ataides, G. Baranes, A. Gu, T. Manovitz, M. Xu, M. Kalinowski, S. Majidy, C. Kokail, N. Maskara, E. C. Trapp, L. M. Stewart, S. Hollerith, H. Zhou, M. J. Gullans, S. F. Yelin, M. Greiner, V. Vuletic, M. Cain, and M. D....

  39. [47]

    A. L. Shaw, Z. Chen, J. Choi, D. K. Mark, P. Scholl, R. Finkelstein, A. Elben, S. Choi, and M. Endres, Nature 628, 71–77 (2024)

  40. [48]

    S. J. Evered, M. Kalinowski, A. A. Geim, T. Manovitz, D. Bluvstein, S. H. Li, N. Maskara, H. Zhou, S. Ebadi, M. Xu, J. Campo, M. Cain, S. Ostermann, S. F. Yelin, S. Sachdev, M. Greiner, V. Vuleti´ c, and M. D. Lukin, Probing topological matter and fermion dy- namics on a neutr...

  41. [49]

    Gonz´ alez-Cuadra, M

    D. Gonz´ alez-Cuadra, M. Hamdan, T. V. Zache, B. Braverman, M. Kornja ca, A. Lukin, S. H. Cant´ u, F. Liu, S.-T. Wang, A. Keesling, M. D. Lukin, P. Zoller, and A. Bylinskii, Nature 642, 321–326 (2025)

  42. [50]

    S. J. Evered, D. Bluvstein, M. Kalinowski, S. Ebadi, T. Manovitz, H. Zhou, S. H. Li, A. A. Geim, T. T. Wang, N. Maskara, H. Levine, G. Semeghini, M. Greiner, V. Vuleti´ c, and M. D. Lukin, Nature 622, 268–272 (2023)

  43. [51]

    Ebadi, A

    S. Ebadi, A. Keesling, M. Cain, T. T. Wang, H. Levine, D. Bluvstein, G. Semeghini, A. Omran, J.-G. Liu, R. Samajdar, X.-Z. Luo, B. Nash, X. Gao, B. Barak, 16 E. Farhi, S. Sachdev, N. Gemelke, L. Zhou, S. Choi, H. Pichler, and et al., Science 376, 1209–1215 (2022)

  44. [52]

    Bluvstein, A

    D. Bluvstein, A. Omran, H. Levine, A. Keesling, G. Se- meghini, S. Ebadi, T. T. Wang, A. A. Michailidis, N. Maskara, W. W. Ho, S. Choi, M. Serbyn, M. Greiner, V. Vuleti´ c, and M. D. Lukin, Science 371, 1355–1359 (2021)

  45. [53]

    Wurtz, S

    J. Wurtz, S. H. Sack, and S.-T. Wang, IEEE Transactions on Quantum Engineering 5, 1–14 (2024)

  46. [54]

    Cerezo, G

    M. Cerezo, G. Verdon, H.-Y. Huang, L. Cincio, and P. J. Coles, Nature Computational Science 2, 567–576 (2022)

  47. [55]

    J. Choi, H. Zhou, H. S. Knowles, R. Landig, S. Choi, and M. D. Lukin, Physical Review X 10, 031002 (2020)

  48. [56]

    Setonix Supercomputer (2023), Pawsey Supercomputing Research Centre, Perth, Western Australia

  49. [57]

    Pokharel, S

    B. Pokharel, S. Srinivasan, G. Quiroz, and B. Boots, Physical Review Research 6, 013187 (2024)

  50. [58]

    D. J. MacKay, Information theory, inference and learning algorithms (Cambridge University Press, 2003)

  51. [59]

    Blume-Kohout, New Journal of Physics 12, 043034 (2010)

    R. Blume-Kohout, New Journal of Physics 12, 043034 (2010)

  52. [60]

    A. P. Dempster, N. M. Laird, and D. B. Rubin, Jour- nal of the Royal Statistical Society Series B: Statistical Methodology 39, 1–22 (1977)

  53. [61]

    Efron, Bootstrap methods: Another look at the jack- knife, in Breakthroughs in Statistics (Springer New York,

    B. Efron, Bootstrap methods: Another look at the jack- knife, in Breakthroughs in Statistics (Springer New York,

  54. [63]

    Bayesian inference Bayesian inference provides a framework to address un- certainties arising through noisy measurement [56, 57]. Given observed data Z and model parameters ϕ, Bayes’ rule states P (ϕ | Z) ∝ P (Z | ϕ) P (ϕ), (A1) where the prior distribution P (ϕ) is the model ...

  55. [64]

    Likelihood under measurement error Measurement error on Aquila is well described by an asymmetric bit-flip channel acting independently on each qubit [23]. The probability of recording a bitstring z given pre-measurement string s is K(z | s) = NY j=1 P (zj | sj), (A2) where j ...

  56. [65]

    uninformative

    Expectation Maximization The expectation–maximization (EM) algorithm is an iterative method for likelihood maximization. The al- gorithm alternates between two steps, which are guar- anteed to monotonically converge to a local optimum of the likelihood [58]. The first is an “E...

  57. [1992]

    p. 569–593. ACKNOWLEDGEMENTS This work was supported by resources provided by the Pawsey Supercomputing Research Centre with fund- ing from the Australian Government and the Govern- ment of Western Australia. The Pawsey Supercomputing Research Centre’s Quantum Supercomputing I...

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.