Pith. sign in

REVIEW 2 major objections 4 minor 75 references

Weakly-Driven Quantum Walks for Memory-Constrained Pauli Channel Learning

T0 review · 2 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read A weakly-driven quantum walk reduces Pauli channel estimation's quantum memory from logarithmic to constant order while preserving its exponential measurement advantage.

desk verdict A plausible constant-memory replacement for Chen–Gong's counting subroutine, but the load-bearing error analysis is heuristic and monotonicity in ε* is asserted, not proven. read the letter →

arxiv 2509.07702 v1 pith:T2J45KMT submitted 2025-09-09 quant-ph

Yuan-Zhuo Wang , Yi-Ran Xiao , Ming-Yang Li , Shengjun Wu , Zeng-Bing Chen This is my paper · ORCID
classification quant-ph
keywords Paulichannelestimationweakly-drivenquantumwalkconstantmemoryhypothesistestingconcatenatedprotocolsmeasurementcomplexitynoisecharacterization
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

The paper tries to establish that the expensive logarithmic-size quantum memory used by a recent Pauli channel estimation protocol can be replaced by a constant number of qubits without losing that protocol's exponential advantage in measurement complexity. The replacement is a single-qubit 'pointer' driven through a weakly-driven quantum walk: repeated small rotations controlled by noisy classical information from the channel accumulate a statistical signal, and a controlled-overwrite channel records the accumulated distinction on a recorder qubit. A double-stage arrangement with reverse overwriting reproduces the error-suppression behavior of the earlier protocol's counting scheme, yielding final survival probabilities S0 ≤ 1/8^n under the null hypothesis and S1 ≥ $e^{{-3}}$ under the alternative. If correct, this removes a key memory bottleneck for characterizing quantum noise, a practical step toward fault-tolerant quantum computing.

What carries the argument

The weakly-driven quantum walk. The pointer is a single qubit whose state lies in the XZ-plane of the Bloch sphere; each step applies a small rotation Ry(+θ) or Ry(-θ) according to the classical probabilistic outcome of a prepared input state, and the small angle θ preserves pointer coherence even though the driving information is classical and random. The controlled-overwrite channel multiplies the recorder's survival probability by the pointer's |0⟩ probability in every round, turning per-round probabilities into a product that is then logarithmically additive. The multi-round varying-step schedule (round i uses i steps) converts the binomial rotation statistics into the approximate log-survival formula whose signal term grows as m³, and the double-stage algorithm appends a reverse overwrite channel and 3n repetitions to convert the inner survival-probability bounds into the outer bounds needed by the Pauli channel estimation framework.

What would settle it

Compute the exact survival probability S(m,θ,ε*) = ∏_{i=1}^{m} (1 − E[sin²(Θ_i/2)]) with Θ_i distributed as (n₊ − n₋)θ for n₊ ∼ Binomial(i, 1/2 + ε*), using the parameters m = c log n/ε² and θ = c′ε²/log n, and check whether the double-stage bounds S0^(2) ≤ 1/8^n and S1^(2) ≥ $e^{{-3}}$ hold for all n and ε. Also scan ε* from ε to 1/2 to see whether S(m,θ,ε*) is monotone decreasing; a single parameter setting where the survival probability increases with ε*, or where the asymptotic bounds fail, would refute the central claim.

Watch

Extended reading notes

Core claim

The central claim is that a quantum walk with deliberately weak single-step driving separates two hypotheses about a Pauli channel's eigenvalue: when the input that controls each step is unbiased, the pointer undergoes purely diffusive dynamics, and when it is biased, the pointer shows drift-diffusion dynamics. A single step applies a rotation Ry(+θ) or Ry(-θ) to the pointer depending on whether the input qubit is in |0⟩ or |1⟩, and after i steps the total rotation angle is a binomial random variable with mean 2iε*θ and variance iθ²(1-4ε*²). Over m rounds with round i using exactly i steps, serial controlled-overwriting makes the recorder's survival probability satisfy approximately −ln S(m,θ,ε*) ≈ m²θ²/8 + m³ε*²θ²/3, so the null decay is quadratic in m while the signal decay is cubic. With m = O(log n/ε²) and θ = O(ε²/log n), the inner subroutine gives S0 ≥ 1/2 and S1 ≤ 1/n; feeding these into a reverse controlled-overwrite channel repeated 3n times gives the outer-stage survival probabilities S0^(2) ≤ 1/8^n and S1^(2) ≥ $e^{{-3}}$, matching the behavior required by the concatenated Pauli channel estimation protocol.

Load-bearing premise

The entire error analysis rests on the approximate formula −ln S(m,θ,ε*) ≈ m²θ²/8 + m³ε*²θ²/3, obtained by treating the binomial walk as Gaussian and keeping only leading Taylor terms, together with the assumption that the recorder's survival probability decreases monotonically as the signal strength grows; the paper gives numerical checks but no rigorous error bounds for the regime m = Θ(log n/ε²), θ = Θ(ε²/log n).

Editorial extensions

If this is right

  • Pauli channel eigenvalues can be estimated with a constant number of memory qubits rather than O(log n), while keeping the measurement-complexity advantage of the concatenated-memory approach.
  • The algorithm requires no intermediate measurements: all parameters are fixed in advance, and the channel queries are serial, making it compatible with the non-adaptive architecture of the baseline protocol.
  • The channel query complexity rises to O(n log² n/ε⁴), a moderate polynomial overhead compared with the original O(n log n/ε²) queries, paid to overcome periodicity with the simple varying-step schedule.
  • The same double-stage logic, with inner survival probabilities S0 ≥ 1/2 and S1 ≤ 1/n followed by 3n reverse overwrites, reproduces the counting scheme's final error behavior S0^(2) ≤ 1/8^n and S1^(2) ≥ e^{-3}.
  • The 'weak driving' principle suggests that pointer coherence can survive driving by high-entropy classical information, opening a route to coherent accumulation of weak classical signals.

Reading between the lines

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

  • A sparse set of O(1) large co-prime step numbers, each of order m, could plausibly resolve the periodicity problem with only O(n log n/ε²) queries, bringing the query complexity back to the original level while retaining the constant-memory advantage; this is an optimization the paper sketches but does not implement.
  • The monotonic decrease of survival probability with signal strength, verified numerically in one parameter regime, suggests the protocol is conservative: its guarantees derived at the threshold ε should only improve for larger deviations, but a rigorous monotonicity proof would place the claim on firmer footing.
  • Since constant memory with an exponential number of queries demands long memory coherence time, the practical value of the result depends on a three-way trade-off among memory qubit count, measurement complexity, and memory coherence time; the paper identifies this trade-off but does not quantify it.
  • The weak-driving mechanism could transfer to other quantum learning and sensing tasks where the signal is weak, classical, and probabilistic, such as estimating small phase shifts or weak noise parameters in resource-constrained settings; this is a speculative extension beyond the paper's explicit scope.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 4 minor

Summary. The paper proposes a hypothesis-testing subroutine based on a 'weakly-driven quantum walk' and integrates it into the Pauli-channel learning framework of Ref. [34]. The key quantitative claim is the approximate survival-probability formula Eq. (4), -ln S ≈ m²θ²/8 + m³ε*²θ²/3, from which the authors derive parameters m = Θ(log n/ε²) and θ = Θ(ε²/log n) that yield inner-loop guarantees S0^(1) ≥ 1/2 and S1^(1) ≤ 1/n. A double-stage algorithm with 3n reverse overwrites then gives S0^(2) ≤ 1/8^n and S1^(2) ≥ e^{-3}, matching the counting-scheme behavior needed by the outer protocol. The authors claim this reduces the quantum memory overhead from O(log n) to O(1) while preserving the exponential measurement-complexity advantage, at the cost of increasing the channel-query complexity from O(n log n/ε²) to O(n log²n/ε⁴).

Significance. If the central estimates can be made rigorous, this is a meaningful advance: it would be the first constant-memory protocol within the Chen-Gong framework to preserve the exponential reduction in measurement rounds, and the 'weak driving' idea is a conceptually interesting bridge between weak measurements and quantum-walk-based sensing. The algorithm is explicit, the numerical checks in Figs. 1 and 2 support the approximate formula in the tested regimes, and the paper honestly identifies where the approximations are expected to be inaccurate. However, the current proof does not supply rigorous error bounds for the asymptotic regime used in the Pauli-channel application, and the reliance on unproved monotonicity in ε* leaves the main resource claim conditional.

major comments (2)
  1. [§IV and Appendix A] Equation (4) is the only quantitative basis for the claimed threshold behavior, but its derivation rests on three uncontrolled approximations: binomial-to-Gaussian, discrete-to-continuous summation, and leading-order Taylor expansions. In the double-stage parameter regime m = Θ(log n/ε²), θ = Θ(ε²/log n), the dropped terms displayed in Eq. (A3), O(iε*²θ², i³ε*²θ⁴), sum to O(ε²) in -ln S, i.e., a constant-order correction independent of n. This can shift S0^(1) by an O(1) factor and violate the stated inequalities S0^(1) ≥ 1/2 and S1^(1) ≤ 1/n unless the constants are chosen with explicit slack and the error terms are bounded. Please provide rigorous error estimates for Eq. (4) in the relevant asymptotic regime, for example via Taylor remainder bounds and Berry-Esseen-type corrections, and verify the inner-loop thresholds with explicit constants.
  2. [Appendix A.2 and Appendix B.3] The monotonicity of the survival probability S(m,θ,ε*) in ε* is asserted in the sentence 'the multi-round, varying-step walk design ensures that the protocol's response is monotonically enhanced with increasing ε*,' but no proof is given, and the appendix explicitly concedes that the approximation can fail when ε* ≫ ε. Figure 2 checks only one parameter tuple (m=85, θ≈0.0277, ε=0.2). Since the encoding in Eq. (7) allows ε* up to 1/2 and Appendix B.3 requires S^(2)_{t*} ≥ e^{-3} for any significant deviation, the outer decision rule needs either a proof that S(m,θ,ε*) ≤ S(m,θ,ε) for all ε* ∈ [ε, 1/2] in the chosen asymptotic regime, or a certified numerical bound over a sufficiently fine grid. Without this, the claimed Type I and II error guarantees for the full range of Pauli-channel deviations do not follow from the analysis presented.
minor comments (4)
  1. [Fig. 1] The x-axis of Fig. 1 ends at 24, but the caption and text report m_opt = 25; the marker for m_opt lies outside the plotted range and should be moved or the axis extended.
  2. [Abstract and §V] The abstract says the protocol 'preserves the exponential advantage in measurement complexity,' but the channel-query complexity increases from O(n log n/ε²) in Ref. [34] to O(n log²n/ε⁴) here; this trade-off is stated later in §V but should be flagged in the abstract or conclusions to avoid overstatement.
  3. [Algorithm 2, line 4] The reset operation 'M1 → |0⟩⟨0|' inside the outer loop is not defined in the resource model; please clarify how this reset is achieved without intermediate measurements and how it interacts with the qubit-reset capability described in §II.
  4. [§IV, Eqs. (5)-(6)] The derivation from Eqs. (5)-(6) to the parameter choices m = Ω(γ/ε²) and θ = O(ε²/γ) is only sketched; writing the asymptotic relations with explicit inequalities and constants would help the reader verify the consistency of the two constraints.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the survival-probability formula is a direct analytic approximation and the walk parameters are derived from target thresholds, not from the predicted survival probabilities.

full rationale

The derivation chain is self-contained. Eq. (4) follows from the specified walk unitary (Eq. (3)) and the controlled-overwrite recurrence p' = <0|rho_P|0> p, via binomial-to-Gaussian approximation, Taylor expansion, and summation of i and i^2 in Appendix A (Eqs. (A1)-(A4)). The parameters m and theta in Eqs. (5)-(6) and in Algorithm 2 (m=O(log n/epsilon^2), theta=O(epsilon^2/log n)) are obtained by solving inequalities for fixed target gamma, epsilon, and n; they are not optimized against observed survival probabilities. Consequently the claims S0^(1) >= 1/2 and S1^(1) <= 1/n are analytic consequences of the approximation, checked by exact binomial numerics in Figs. 1-2, rather than renamed fits. The outer Pauli-channel protocol is taken from the external Ref. [34] (PRX Quantum 6, 020323 (2025)), with no overlapping authors, so the 'constant memory while preserving poly(n) advantage' claim rests on an independent framework, not on a self-citation chain. Self-citations [44,46,55-57] are background for quantum walks and weak measurement and are not load-bearing. Appendix A.2 explicitly concedes the approximation can fail for epsilon* >> epsilon and invokes monotonicity of S(m,theta,epsilon*) in epsilon* without proof; this is an unproved mathematical assertion and a correctness risk, but it is not circular because the monotonicity claim is not identical to Eq. (4) nor to the parameter definitions. Hence score 0.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

No data-fitted parameters or new physical entities are introduced. The algorithm parameters m and theta are chosen from analytic constraints, and the weakly-driven quantum walk is an algorithmic mechanism rather than a postulated physical degree of freedom. The main burden is the unproved accuracy of the approximate survival-probability formula and the monotonicity assumption used to extend the analysis to strong signals.

free parameters (2)
  • m (number of rounds) = m = Omega(gamma/epsilon^2) standalone; m = O(log n/epsilon^2) in double-stage
    Chosen by hand to satisfy the constraints m theta = O(1) and m^3 epsilon^2 theta^2 = Omega(gamma). The guarantees depend on unspecified constants in the O-notation.
  • theta (single-step rotation angle) = theta = O(epsilon^2/gamma) standalone; theta = O(epsilon^2/log n) in double-stage
    Chosen from the same constraints; the small-angle Taylor expansion used in Eq. (4) assumes theta and i epsilon* theta are small over the relevant rounds.
assumptions (5)
  • domain assumption Idealized resources: noiseless ancilla qubits, universal error-free gates, deterministic measurement-free reset, and serially independent channel queries.
    These computational resources are assumed in Sec. II and are essential for the constant-memory claim. The reset capability, in particular, is used in every round of Algorithms 1 and 2.
  • ad hoc to paper The Gaussian approximation of the binomial rotation count and the leading-order Taylor expansions in Appendix A are accurate enough in the asymptotic parameter regime.
    Eq. (4) and the error analysis in Appendix A.2 rely on this assumption. Only two numerical checks are given, and no rigorous error bound is proved.
  • ad hoc to paper The survival probability S(m, theta, epsilon*) is monotonically decreasing in epsilon* for epsilon* >= epsilon.
    Used in Sec. IV and Appendix A.2 to extend the worst-case analysis at epsilon* = epsilon to all stronger signals. The paper argues the varying-step design prevents harmful periodicity but proves this only numerically in Fig. 2.
  • standard math The POVM encoding of lambda_t - lambda_hat_t can be implemented as a measurement-free quantum channel via Stinespring dilation with two reusable ancilla qubits.
    Standard dilation theory is invoked in Appendix B.1; the ancilla reset requirement is supplied by the assumed reset capability.
  • domain assumption The protocol framework of Ref. [34], including its outer loop over 4^n - 1 eigenvalues and its use of 3n serial invocations, remains valid when the counting scheme is replaced by the presented subroutine.
    The paper assumes the replacement preserves the outer protocol's error behavior; the analysis in Appendix B.3 checks the two relevant hypotheses but does not re-derive the full framework.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Weakly-Driven Quantum Walks for Memory-Constrained Pauli Channel Learning." pith.science (2026). https://pith.science/paper/T2J45KMT

@misc{pith2026250907702,
  author       = {Pith},
  title        = {Pith review of: Weakly-Driven Quantum Walks for Memory-Constrained Pauli Channel Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/T2J45KMT}},
  note         = {Machine review of arXiv:2509.07702}
}
read the original abstract

Accurate characterization of quantum noise, exemplified by the Pauli channel, is a cornerstone for building fault-tolerant quantum computers. A recent protocol (PRX Quantum 6, 020323 (2025)) combining channel concatenation and quantum memory has achieved an exponential reduction in measurement complexity for Pauli channel estimation. This efficiency, however, hinges on using logarithmic quantum memory to suppress hypothesis test errors. In this work, we introduce a mechanism termed the ``weakly-driven quantum walk'' to mitigate the demand for high-quality quantum memory. By exploiting the distinct dynamical properties of quantum walks under biased versus unbiased driving, our algorithm lowers the quantum memory overhead to a constant order while preserving the exponential advantage in measurement complexity. By analogy with weak measurement, our introduced concept of ``weak driving'' preserves pointer coherence even when driven by classical probabilistic information, a principle that may inspire new approaches to similar quantum algorithm design and quantum sensing of weak signals in resource-constrained scenarios.

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

75 extracted references · 70 canonical work pages

  1. [34]

    van den Berg, Z

    E. van den Berg, Z. K. Minev, A. Kandala, and K. Temme, Probabilistic Error Cancellation with Sparse Pauli–Lindblad Models on Noisy Quantum Processors, Nat. Phys. 19, 1116 (2023)

    work page 2023

  2. [1]

    Constant quantum memory: The algorithm can use O(1) noiseless and decoherence-free ancilla qubits, which serve as quantum memory to store intermediate information or as working qubits for computation

    work page

  3. [2]

    Universal quantum gates: The algorithm can employ universal and error-free quantum gates in its circuit

    work page

  4. [3]

    Qubit reset capability: The algorithm can de- terministically reset a specified qubit to the |0⟩ state within the circuit via a measurement-free quantum channel

    work page

  5. [4]

    controlled- overwrite channel,

    Concatenated channel queries: The algorithm can serially query a quantum channel that outputs the state ρin(ε∗) multiple times, with each output being independent of the others. III. DESIGN PHILOSOPHY AND ALGORITHMIC COMPONENTS A. Strongly-Driven and W eakly-Driven Quantum W alks Given our capability for concatenated channel queries, we can subject a quan...

    work page

  6. [5]

    , m}) consists of an i-step quantum walk driven by the in- put state ρin(ε∗)

    Derivation of the Survival Probability In our algorithm, the i-th round (for i ∈ { 1, . . . , m}) consists of an i-step quantum walk driven by the in- put state ρin(ε∗). Let n+ be the number of steps with a + θ rotation and n− be the number of steps with a −θ rotation, such that the total number of steps is i = n+ + n−. The final total rotation angle for t...

    work page

  7. [6]

    Justification for the Validity of the Approximations The approximations used to obtain the above results include: approximating the discrete binomial distribu- tion with a continuous Gaussian distribution, applying a continuous approximation to the summation formulas, and performing Taylor expansions on the trigonometric, exponential, and logarithmic funct...

    work page

  8. [7]

    Deviation-Encoding Channel To encode the difference between an unknown non- trivial eigenvalue λt of an n-qubit Pauli channel and its hypothesized value ˆλt (where t ∈ { 1, . . . ,4n − 1} and λt, ˆλt ∈ [−1, 1]) onto a sample ancilla qubit in the form of our required input state, ρin = (1 /2 + ε∗) |0⟩ ⟨ 0| + (1/2 − ε∗) |1⟩ ⟨ 1|, we first define a parameterize...

    work page

Show all 75 references
  1. [8]

    This imposes the constraint: c1(1+ ˆλt)+ c2(1 − ˆλt) = 1

    Zero-deviation calibration: Under the null hy- pothesis, i.e., λt = ˆλt, the probability must be ex- actly 1/2. This imposes the constraint: c1(1+ ˆλt)+ c2(1 − ˆλt) = 1

  2. [9]

    Solving this constrained optimization problem yields the optimal coefficients

    Sensitivity maximization: Subject to the cali- bration constraint, we must maximize the local re- sponse of the probability to changes in λt, which means maximizing the absolute value of its deriva- tive: | ∂p ∂λt | = | c1−c2 2 |. Solving this constrained optimization problem ...

  3. [10]

    As can be seen, its logic is the inverse of the controlled- overwrite channel Cwrite: if M1 is in the |0⟩ state, M2 is overwritten to |1⟩ ; otherwise, its state remains un- changed

    The Double-Stage Algorithm Implementing this algorithm requires the introduction of the reverse controlled-overwrite channel, Cr-write, which is described by a set of three Kraus operators: K1 = |1⟩ ⟨ 1|M1 ⊗ IM2 , K0,0 = |0⟩ ⟨ 0|M1 ⊗ |1⟩ ⟨ 0|M2 , K0,1 = |0⟩ ⟨ 0|M1 ⊗ |1⟩ ⟨ 1|M2...

  4. [11]

    First stage: We first run the m-round weakly- driven quantum walk subroutine, with its result recorded on the first recorder qubit, M1. By se- lecting the parameters m = O(log n/ε2) and θ = O(ε2/ log n), we can ensure that the survival prob- ability of M1, denoted S(1), satisfies...

  5. [12]

    The survival probability of M2 in a single it- eration becomes: 1 − S(1) 0 ≤ 1/2 under H0, and 1 − S(1) 1 ≥ 1 − 1/n under H1

    Second stage: Next, through the Cr-write opera- tion acting on M1 and M2, we logically invert the probability information from M1 and transfer it to M2. The survival probability of M2 in a single it- eration becomes: 1 − S(1) 0 ≤ 1/2 under H0, and 1 − S(1) 1 ≥ 1 − 1/n under H1...

  6. [13]

    [ 34], the output of the double- stage algorithm, the second recorder qubit M2, is used as a control qubit during the tests of the 4 n − 1 eigenvalues

    Performance Analysis in Pauli Channel Estimation In the protocol of Ref. [ 34], the output of the double- stage algorithm, the second recorder qubit M2, is used as a control qubit during the tests of the 4 n − 1 eigenvalues. It acts via another reverse controlled-overwrite cha...

  7. [14]

    Aharonov, J

    D. Aharonov, J. Cotler, and X.-L. Qi, Quantum Algo- rithmic Measurement, Nat. Commun. 13, 887 (2022)

    2022

  8. [15]

    Bubeck, S

    S. Bubeck, S. Chen, and J. Li, Entanglement Is Necessary for Optimal Quantum Property Testing, in 2020 IEEE 61st Annual Symposium on Foundations of Computer S cience (FOCS) (2020) pp. 692–703

    2020

  9. [16]

    S. Chen, J. Cotler, H.-Y. Huang, and J. Li, The Com- plexity of NISQ, Nat. Commun. 14, 6001 (2023)

    2023

  10. [17]

    S. Chen, J. Cotler, H.-Y. Huang, and J. Li, Exponential Separations Between Learning With and Without Quantum Memory, in 2021 IEEE 62nd Annual Symposium on Foundations of Computer S cience (FOCS) (2022) pp. 574–585

    2022

  11. [18]

    S. Chen, B. Huang, J. Li, A. Liu, and M. Sellke, When Does Adaptivity Help for Quantum State Learning?, in 2023 IEEE 64th Annual Symposium on Foundations of Computer S cience (FOCS) (2023) pp. 391–404

    2023

  12. [19]

    Huang, R

    H.-Y. Huang, R. Kueng, and J. Preskill, Information- Theoretic Bounds on Quantum Advantage in Machine Learning, Phys. Rev. Lett. 126, 190505 (2021)

    2021

  13. [20]

    Huang, M

    H.-Y. Huang, M. Broughton, J. Cotler, S. Chen, J. Li, M. Mohseni, H. Neven, R. Babbush, R. Kueng, J. Preskill, and J. R. McClean, Quantum Advantage in Learning from Experiments, Science 376, 1182 (2022)

    2022

  14. [21]

    P. W. Shor, Algorithms for Quantum Compu- tation: Discrete Logarithms and Factoring, in Proceedings 35th Annual Symposium on Foundations of Comput er Science (1994) pp. 124–134

    1994

  15. [22]

    Aaronson and A

    S. Aaronson and A. Arkhipov, The Com- putational Complexity of Linear Optics, in Proceedings of the 43rd Annual ACM Symposium on Theory of Com puting (2011) pp. 333–342

    2011

  16. [23]

    A. W. Harrow and A. Montanaro, Quantum Computa- tional Supremacy, Nature 549, 203 (2017)

    2017

  17. [24]

    Arute, K

    F. Arute, K. Arya, R. Babbush, D. Bacon, J. C. Bardin, et al. , Quantum Supremacy Using a Programmable Su- perconducting Processor, Nature 574, 505 (2019)

    2019

  18. [25]

    Zhong, H

    H.-S. Zhong, H. Wang, Y.-H. Deng, M.-C. Chen, L.-C. Peng, et al. , Quantum Computational Advantage Using Photons, Science 370, 1460 (2020)

    2020

  19. [26]

    P. W. Shor, Fault-Tolerant Quantum Computation, in Proceedings of 37th Conference on Foundations of Computer S cience (1996) pp. 56–65

    1996

  20. [27]

    Gottesman, Theory of Fault-Tolerant Quantum Com- putation, Phys

    D. Gottesman, Theory of Fault-Tolerant Quantum Com- putation, Phys. Rev. A 57, 127 (1998)

    1998

  21. [28]

    Singh, C

    K. Singh, C. E. Bradley, S. Anand, V. Ramesh, R. White, and H. Bernien, Mid-Circuit Correction of Correlated Phase Errors Using an Array of Spectator Qubits, Science 380, 1265 (2023)

    2023

  22. [29]

    J. J. Wallman and J. Emerson, Noise Tailoring for Scal- able Quantum Computation via Randomized Compiling, Phys. Rev. A 94, 052325 (2016)

    2016

  23. [30]

    Hashim, R

    A. Hashim, R. K. Naik, A. Morvan, J.-L. Ville, B. Mitchell, J. M. Kreikebaum, M. Davis, E. Smith, C. Iancu, K. P. O’Brien, I. Hincks, J. J. Wall- man, J. Emerson, and I. Siddiqi, Randomized Compiling for Scalable Quantum Computing on a Noisy Superconducting Quantum Processor, ...

    2021

  24. [31]

    Erhard, J

    A. Erhard, J. J. Wallman, L. Postler, M. Meth, R. Stricker, E. A. Martinez, P. Schindler, T. Monz, J. Emerson, and R. Blatt, Characterizing Large- Scale Quantum Computers via Cycle Benchmarking, Nat. Commun. 10, 5347 (2019)

    2019

  25. [32]

    Harper, S

    R. Harper, S. T. Flammia, and J. J. Wallman, Efficient Learning of Quantum Noise, Nat. Phys. 16, 1184 (2020)

    2020

  26. [33]

    Carignan-Dugas, D

    A. Carignan-Dugas, D. Dahlen, I. Hincks, E. Ospadov, S. J. Beale, S. Ferracin, J. Skanes-Norman, J. Emerson, and J. J. Wallman, The Error Reconstruction and Com- piled Calibration of Quantum Computing Cycles (2023), arXiv:2303.17714 [quant-ph]

    2023 arXiv

  27. [35]

    Ferracin, A

    S. Ferracin, A. Hashim, J.-L. Ville, R. Naik, A. Carignan- Dugas, H. Qassim, A. Morvan, D. I. Santiago, I. Siddiqi, and J. J. Wallman, Efficiently Improv- ing the Performance of Noisy Quantum Computers, Quantum 8, 1410 (2024) , arXiv:2201.10672

    2024 arXiv

  28. [36]

    Y. Kim, A. Eddins, S. Anand, K. X. Wei, E. van den Berg, S. Rosenblatt, H. Nayfeh, Y. Wu, M. Za- letel, K. Temme, and A. Kandala, Evidence for the Utility of Quantum Computing before Fault Tolerance, 12 Nature 618, 500 (2023)

    2023

  29. [37]

    D. K. Tuckett, S. D. Bartlett, and S. T. Flammia, Ul- trahigh Error Threshold for Surface Codes with Biased Noise, Phys. Rev. Lett. 120, 050505 (2018)

    2018

  30. [38]

    Fujiwara and H

    A. Fujiwara and H. Imai, Quantum Parame- ter Estimation of a Generalized Pauli Channel, J. Phys. A: Math. Gen. 36, 8093 (2003)

    2003

  31. [39]

    Chiuri, V

    A. Chiuri, V. Rosati, G. Vallone, S. P´ adua, H. Imai, S. Giacomini, C. Macchiavello, and P. Mataloni, Experimental Realization of Opti- mal Noise Estimation for a General Pauli Channel, Phys. Rev. Lett. 107, 253602 (2011)

    2011

  32. [40]

    S. T. Flammia and J. J. Wallman, Ef- ficient Estimation of Pauli Channels, ACM Trans. Quantum Comput. 1, 1 (2020) , arXiv:1907.12976

    2020 arXiv

  33. [41]

    S. T. Flammia and R. O’Donnell, Pauli Error Estima- tion via Population Recovery, Quantum 5, 549 (2021) , arXiv:2105.02885

    2021 arXiv

  34. [42]

    Harper, W

    R. Harper, W. Yu, and S. T. Flammia, Fast Estimation of Sparse Quantum Noise, PRX Quantum 2, 010322 (2021)

    2021

  35. [43]

    S. Chen, S. Zhou, A. Seif, and L. Jiang, Quan- tum Advantages for Pauli Channel Estimation, Phys. Rev. A 105, 032435 (2022) , arXiv:2108.08488

    2022 arXiv

  36. [44]

    S. Chen, Y. Liu, M. Otten, A. Seif, B. Feffer- man, and L. Jiang, The Learnability of Pauli Noise, Nat. Commun. 14, 52 (2023)

    2023

  37. [45]

    S. Chen, C. Oh, S. Zhou, H.-Y. Huang, and L. Jiang, Tight Bounds on Pauli Channel Learning with- out Entanglement, Phys. Rev. Lett. 132, 180805 (2024) , arXiv:2309.13461

    2024 arXiv

  38. [46]

    Fawzi, A

    O. Fawzi, A. Oufkir, and D. S. Fran¸ ca, Lower Bounds on Learning Pauli Channels with Individual Mea- surements, IEEE Trans. Inf. Theory 71, 2642 (2025) , arXiv:2301.09192

    2025 arXiv

  39. [47]

    Chen and W

    S. Chen and W. Gong, Efficient Pauli channel estimation with Logarithmic quantum memory, PRX Quantum 6, 020323 (2025)

    2025

  40. [48]

    A. I. Lvovsky, B. C. Sanders, and W. Tittel, Optical Quantum Memory, Nat. Photonics 3, 706 (2009)

    2009

  41. [49]

    H´ etet, J

    G. H´ etet, J. J. Longdell, M. J. Sellars, P. K. Lam, and B. C. Buchler, Multimodal Properties and Dynamics of Gradient Echo Quantum Memory, Phys. Rev. Lett. 101, 203601 (2008)

    2008

  42. [50]

    K. F. Reim, P. Michelberger, K. C. Lee, J. Nunn, N. K. Langford, and I. A. Walmsley, Single-Photon- Level Quantum Memory at Room Temperature, Phys. Rev. Lett. 107, 053603 (2011)

    2011

  43. [51]

    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

  44. [52]

    Z. Yu, Z. Wu, X. Li, X. Feng, W. Huang, K. Zhang, C.-H. Yuan, W. Zhang, and L. Q. Chen, Interferometry-Integrated Noise-Immune Quan- tum Memory, Phys. Rev. Lett. 131, 150804 (2023)

    2023

  45. [53]

    S. E. Thomas, L. Wagner, R. Joos, R. Sittig, C. Nawrath, P. Burdekin, I. M. d. B. Wenniger, M. J. Rasiah, T. Huber-Loyola, S. Sagona-Stophel, S. H¨ ofling, M. Jet- ter, P. Michler, I. A. Walmsley, S. L. Portalupi, and P. M. Ledingham, Deterministic Storage and Retrieval of Tele...

    2024

  46. [54]

    Aharonov, L

    Y. Aharonov, L. Davidovich, and N. Zagury, Quantum Random Walks, Phys. Rev. A 48, 1687 (1993)

    1993

  47. [55]

    Aharonov, A

    D. Aharonov, A. Ambainis, J. Kempe, and U. Vazirani, Quantum Walks on Graphs, in Proceedings of the 33rd Annual ACM Symposium on Theory of Com puting (2001) pp. 50–59

    2001

  48. [56]

    A. M. Childs, R. Cleve, E. Deotto, E. Farhi, S. Gutmann, and D. A. Spielman, Exponential Algorithmic Speedup by a Quantum Walk, in Proceedings of the 35th Annual ACM Symposium on Theory of Com puting (2003) pp. 59–68

    2003

  49. [57]

    C. Lyu, L. Yu, and S. Wu, Localization in Quantum Walks on a Honeycomb Network, Phys. Rev. A 92, 052305 (2015)

    2015

  50. [58]

    M. Gong, S. Wang, C. Zha, M.-C. Chen, H.-L. Huang, et al. , Quantum Walks on a Programmable Two-Dimensional 62-Qubit Superconducting Processor, Science 372, 948 (2021)

    2021

  51. [59]

    Wang, J.-Y

    L.-J. Wang, J.-Y. Lin, and S. Wu, Implementation of Quantum Stochastic Walks for Function Approxima- tion, Two-Dimensional Data Classification, and Sequence Classification, Phys. Rev. Res. 4, 023058 (2022)

    2022

  52. [60]

    Attal, F

    S. Attal, F. Petruccione, C. Sabot, and I. Sinayskiy, Open Quantum Random Walks, J. Stat. Phys. 147, 832 (2012)

    2012

  53. [61]

    Sinayskiy and F

    I. Sinayskiy and F. Petruccione, Open Quantum Walks, Eur. Phys. J. Spec. Top. 227, 1869 (2019)

    2019

  54. [62]

    C. S. Hamilton, R. Kruse, L. Sansoni, C. Sil- berhorn, and I. Jex, Driven Quantum Walks, Phys. Rev. Lett. 113, 083602 (2014)

    2014

  55. [63]

    P. Held, M. Engelkemeier, S. De, S. Barkhofen, J. Sper- ling, and C. Silberhorn, Driven Gaussian Quantum Walks, Phys. Rev. A 105, 042210 (2022)

    2022

  56. [64]

    M. B. Mensky, Quantum Restrictions for Continuous Ob- servation of an Oscillator, Phys. Rev. D 20, 384 (1979)

    1979

  57. [65]

    Aharonov, D

    Y. Aharonov, D. Z. Albert, and L. Vaidman, How the Result of a Measurement of a Component of the Spin of a Spin-1/2 Particle Can Turn out to Be 100, Phys. Rev. Lett. 60, 1351 (1988)

    1988

  58. [66]

    V. P. Belavkin, Quantum Continual Measure- ments and a Posteriori Collapse on CCR, Commun. Math. Phys. 146, 611 (1992)

    1992

  59. [67]

    Hosten and P

    O. Hosten and P. Kwiat, Observation of the Spin Hall Effect of Light via Weak Measurements, Science 319, 787 (2008)

    2008

  60. [68]

    Wu and Y

    S. Wu and Y. Li, Weak Measurements be- yond the Aharonov-Albert-Vaidman Formalism, Phys. Rev. A 83, 052106 (2011)

    2011

  61. [69]

    S. Pang, S. Wu, and Z.-B. Chen, Weak Measure- ment with Orthogonal Preselection and Postselection, Phys. Rev. A 86, 022112 (2012)

    2012

  62. [70]

    Wu, State Tomography via Weak Measurements, Sci

    S. Wu, State Tomography via Weak Measurements, Sci. Rep. 3, 1193 (2013)

    2013

  63. [71]

    P. O. Boykin, T. Mor, V. Roychowdhury, F. Vatan, and R. Vrijen, Algorithmic Cool- ing and Scalable NMR Quantum Computers, Proc. Natl. Acad. Sci. U.S.A. 99, 3388 (2002)

    2002

  64. [72]

    Geerlings, Z

    K. Geerlings, Z. Leghtas, I. M. Pop, S. Shankar, L. Frun- zio, R. J. Schoelkopf, M. Mirrahimi, and M. H. Devoret, Demonstrating a Driven Reset Protocol for a Supercon- ducting Qubit, Phys. Rev. Lett. 110, 120501 (2013)

    2013

  65. [73]

    Magnard, P

    P. Magnard, P. Kurpiers, B. Royer, T. Walter, J.- C. Besse, S. Gasparinetti, M. Pechal, J. Heinsoo, S. Storz, A. Blais, and A. Wallraff, Fast and Uncondi- tional All-Microwave Reset of a Superconducting Qubit, 13 Phys. Rev. Lett. 121, 060502 (2018)

    2018

  66. [74]

    ´A. M. Alhambra, M. Lostaglio, and C. Perry, Heat- Bath Algorithmic Cooling with Optimal Thermalization Strategies, Quantum 3, 188 (2019)

    2019

  67. [75]

    M. A. Aamir, P. Jamet Suria, J. A. Mar ´ ın Guzm´ an, C. Castillo-Moreno, J. M. Epstein, N. Yunger Halpern, and S. Gasparinetti, Thermally Driven Quantum Refrig- erator Autonomously Resets a Superconducting Qubit, Nat. Phys. 21, 318 (2025)

    2025

Pith tools

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