Pith. sign in

REVIEW 3 major objections 6 minor 1 cited by

Performance analysis of GKP error correction

T0 review · 3 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Knill-style GKP error correction with a qunaught-state Bell resource outperforms all other variants and subsumes Steane correction as a non-optimal special case.

desk verdict Solid analytical comparison of GKP error-correction circuits, but the 'superior without trade-off' claim only holds under a worst-case quadrature metric and ignores biased-noise and qunaught preparation costs. read the letter →

arxiv 2505.14775 v1 pith:Y42HMPNK submitted 2025-05-20 quant-ph

Frederik K. Marqversen , Janus H. Wesenberg , Nikolaj T. Zinner , Ulrik L. Andersen This is my paper · ORCID
classification quant-ph PACS 03.67.Pp
keywords GKPerrorcorrectionKnillSteanequnaughtstatessqueezingquantumteleportationbosoniccontinuous-variablecomputing
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

Gottesman-Kitaev-Preskill (GKP) codes correct continuous noise through measurement-based error correction, and the two standard routes are teleportation-based Knill correction and sequential-measurement Steane correction. This paper argues that these are not two separate methods: the Steane scheme is exactly Knill correction with a particular Bell resource, the one produced by the standard qubit-style preparation circuit. When the Bell resource is instead prepared from two qunaught states (empty GKP states carrying no logical information) through a beam splitter, the corrected state keeps both quadratures squeezed at the ancilla width $\Delta^2$ and shows no displacement errors correlated with the input. Under the same resource assumptions, the qunaught-based Knill circuit is therefore both better and simpler, requiring only passive linear optics and homodyne measurement, and the paper backs this with closed-form expressions for post-correction squeezing and displacement errors.

What carries the argument

The load-bearing object is the projection operator $\Pi_\Delta=(1/\sqrt{2\pi})\int dq'\,dq''\,\langle q',q''|\Phi^+\rangle_\Delta |q''\rangle\langle q'|$, which factors all noise correction away from the measurement outcome: $K(s)=\Pi_\Delta D(-s)$. The proof works by showing that the two entangling choices, controlled displacement and a balanced beam splitter, produce exactly this projector, and that the Steane circuit reduces to the same projector with the Bell state $|\Phi^+\rangle_{0,\Delta}$ built by the standard qubit preparation circuit. Performance is read off from a Gaussian-integral lemma applied to the spike decompositions: the output is again a Gaussian spike train whose common width and peak positions are explicit rational functions of the input and resource precision matrices. The qunaught state $|\emptyset\rangle_L=\sum_n |n\sqrt{2\pi}\rangle$, entangled by a beam splitter, yields a resource with diagonal precision matrix $Q=\Delta^{-2}I$, which is what makes the output symmetric and displacement-free.

What would settle it

Prepare the same noisy GKP input at several values of $\Delta$, apply the qunaught/beam-splitter Knill circuit and the standard Bell circuit, and reconstruct the output Wigner functions: the qunaught output must show equal $q$- and $p$-peak widths equal to $\Delta^2$ and lattice positions independent of the input displacement, while the standard output must show the asymmetric widths of eqs. (47) and (49). Finding a qunaught output with $\sigma^2_{\text{out}}\neq\Delta^2$, or a displacement correlated with the input, would refute the central parametrization; alternatively, a surface-GKP threshold simulation with biased noise that makes the Steane variant win at equal $\Delta$ would refute the 'without trade-off' conclusion.

Watch

Extended reading notes

Core claim

The central result is an exact factorization of GKP error correction: every Knill-type circuit acts as $K(s)=\Pi_\Delta D(-s)$, a displacement by the measured syndrome followed by a projection $\Pi_\Delta$ that depends only on the entangled resource state, not on the input. Within a Gaussian-spike approximation in which input and resource states are sums of identical-width Gaussian peaks, the projection is evaluated in closed form: output spikes have width $\sigma^2_{\text{out}}=(Q_{11}+Q_{\text{in}})/(\det Q+Q_{\text{in}}Q_{22})$ and peaks shifted by $(Q_{\text{in}}Q_{12}/(\det Q+Q_{\text{in}}Q_{22}))(\mu_1-\mu_{\text{in}})$. For a Bell state made from qunaught states through a beam splitter, the resource precision matrix is $Q=\Delta^{-2}I$, so $\sigma^2_{\text{out}}=\Delta^2$ in both quadratures and the displacement term vanishes; for the standard Bell state the same formula yields the asymmetric widths of eqs. (47) and (49), one quadrature widened beyond $\Delta^2$ and the other squeezed below it, with lattice positions correlated with previous input displacements. The paper therefore establishes the Steane scheme as a non-optimal special case of the Knill scheme and identifies the qunaught/beam-splitter setting as the one that preserves squeezing and lattice alignment with the simplest linear-optics implementation.

Load-bearing premise

The ranking rests on giving every resource state the same peak width $\Delta$, measuring quality by the product of the two output variances, and modeling all states as Gaussian-spike mixtures; if qunaught states cost more to make or biased noise rewards an asymmetric output, the ordering can flip.

Editorial extensions

If this is right

  • Steane-type GKP correction inherits its full performance profile from Knill correction with the standard Bell preparation, so it need not be designed, simulated, or optimized as a separate scheme.
  • With qunaught-based Bell states, a corrected GKP state has peak width $\Delta^2$ in both quadratures regardless of the input width, and its lattice positions do not depend on prior displacement errors.
  • The standard (Steane-equivalent) Bell state gives one quadrature an output width larger than $\Delta^2$ and the other smaller, with a product that grows whenever $\sigma^2_{\text{in},q}\,\sigma^2_{\text{in},p}\ge\Delta^4$, meaning typical inputs lose total GKP squeezing.
  • Because the beam-splitter version requires only passive linear optics and homodyne detection, the better-performing option is also the experimentally lighter one in photonic hardware, provided qunaught states are available.
  • The general output-width formula applies beyond the photon-number-dampening model used for the comparisons, since the paper cites equivalence among common GKP noise approximations.

Reading between the lines

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

  • A decoder that faces biased noise could prefer the Steane-type asymmetric output on purpose: it squeezes one quadrature at the expense of the other, which may match a channel whose dominant errors are one-sided; the paper's ranking assumes symmetric resources and judges by the variance product.
  • The 'no trade-off' statement moves the cost of qunaught preparation off the books. Qunaught states are non-Gaussian and require a nonlinear or measurement-based source, so a fair resource count must include that preparation step before declaring the linear-optics circuit strictly simpler.
  • The factorization $K(s)=\Pi_\Delta D(-s)$ suggests a decoder-side separation: syndrome estimation and ancilla-quality estimation can be handled independently, and this structure could be exploited in concatenated surface-GKP decoding.
  • A direct experimental test would compare the two Bell preparations at equal resource squeezing: if the corrected Wigner function from the qunaught circuit shows residual displacement correlated with the input displacement, the paper's displacement-free property is violated.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. The paper analyzes continuous-variable GKP error correction circuits, focusing on two main approaches: Knill-type (teleportation-based) error correction and Steane-type (QND-based) error correction. It proves that two different implementations of the Knill approach (one using a controlled-X gate, one using a beam splitter) are mathematically equivalent, and that the Steane approach is a special case of the Knill approach when the ancilla Bell state is prepared by the standard qubit circuit. The central analytical result is a closed-form expression, Eq. (33), for the post-correction Gaussian spike width as a function of the input state's spike width and the ancilla Bell state's precision matrix. This formula is then used to compare two ancilla preparations: a qunaught-state-based Bell state, which yields symmetric output widths Δ² in both quadratures, and the standard qubit-based Bell state, which yields an asymmetric output with one quadrature amplified and the other squeezed. The paper concludes that the qunaught-based Knill scheme achieves superior GKP squeezing and is also the simplest to implement experimentally, without trade-off.

Significance. If the comparison is properly scoped, the paper makes a useful contribution to the continuous-variable quantum error correction literature. The explicit derivation of the post-correction spike width for a general Gaussian resource state, the analytic proof that the Steane circuit is a special case of the Knill circuit, and the numerical validation with Strawberry Fields (maximum error below 10^-8) are valuable and give confidence in the circuit equivalences. The publicly available code for reproducing the figures is a further strength. However, the headline claim of unconditional superiority of the qunaught-based scheme is not supported by the paper's own metric, and the experimental simplicity claim omits the cost of preparing qunaught states. With a re-scoped conclusion and an explicit performance metric, this work would be of interest to researchers working on GKP-based fault tolerance and photonic implementations.

major comments (3)
  1. [Section 5.2, Eq. (50), and Section 6] The conclusion that the qunaught-based Knill scheme achieves 'superior GKP squeezing ... without trade-off' is not supported by the paper's own metric. Under the comparison premise used in the abstract and conclusion, namely the same squeezing parameter Δ for all resource states and a symmetric input noise model as in Eq. (36), the input width equals the ancilla width, so x=y=1. Equation (50) then gives exp(2r_q)exp(2r_p)=1. Equations (47) and (49) yield σ_q^2=3Δ^2/2 and σ_p^2=2Δ^2/3 for the standard/Steane scheme, whose product is Δ^4, identical to the qunaught outcome. Thus qunaught is superior only in the maximum (worst-case) quadrature variance, a metric that is not stated or defended in the paper. Moreover, under biased Gaussian displacement noise the tighter p quadrature of the standard/Steane output can be an advantage, as suggested by Ref. [20] on biased-noise GKP codes. The unconditional 'without trade-off' ranking is therefore not justified. The central claim should be restated as a symmetry or worst-case-quadrature advantage under an explicit metric, or the metric should be defined and justified.
  2. [Section 6 and Abstract] The claim that the qunaught-based approach is 'simplest to realize experimentally' and requires 'only linear optics' is scoped to the entangling circuit only. The resource state |∅⟩_Δ with √(2π) spacing is a non-Gaussian qunaught GKP state; preparing it at the same squeezing Δ as the standard |0⟩_Δ ancilla still requires non-Gaussian operations or non-Gaussian resource states, and the paper does not quantify or compare this preparation overhead. As a result, the 'without trade-off' claim also ignores a relevant resource cost. Please either include the qunaught preparation cost in the comparison or explicitly limit the simplicity claim to the entangling operation rather than the full scheme.
  3. [Section 4 and Eq. (33)] The central formula for the corrected spike width, Eq. (33), is derived under the assumption that both the input state and the ancilla Bell state are exact superpositions of identical Gaussian spikes with a common covariance matrix, as stated in Eqs. (28)-(29). Realistic finite-energy GKP states have a global envelope and overlapping peaks; these effects are outside the Gaussian-spike ansatz. Since Eqs. (40), (47), and (49) are the quantitative basis for the performance comparison, the paper should either demonstrate numerically, using the provided Strawberry Fields code, that these width formulas remain accurate at finite squeezing levels relevant to fault tolerance (around 10 dB), or clearly state that the comparison is an ideal-spike-model result. Without this, the abstract's claim of an 'analytical expression for the post-correction GKP squeezing' overstates what is derived for the idealized model.
minor comments (6)
  1. [Section 3, after Eq. (18)] There is a typo: 'the the projection operator' should read 'the projection operator'.
  2. [Eq. (50)] The rightmost expression with z^2 is needlessly opaque; rewriting it in the direct form 1 + (x^2 y^2 - 1)/((1+x^2)(2+y^2)) would make the condition for a net gain or loss of total squeezing transparent.
  3. [Section 5.2] The sentence 'the above covariance is equivalent to simply swapping the two modes of the Bell state' is unclear; please specify the mode-swap operation explicitly or provide the transformed precision matrix.
  4. [Figure 4 caption] The caption refers to 'circuit 37 and circuit 41'; since these are numbered equations, it should read 'Eq. (37) and Eq. (41)' or 'circuit (37) and circuit (41)'.
  5. [Introduction and Section 6] The phrases 'optimal choice of the entangled resource state' and 'non-optimal special case' are not established by the analysis, which only compares two specific Bell-state preparations. Please weaken these to refer to 'the particular resource states considered here' unless an optimization over all possible states is performed.
  6. [Section 2, Eq. (4)] The qunaught state is described as a 'squeezed logical state'; it would help to explicitly explain why the √(2π) spacing implies that it encodes no logical information, since this is central to the later discussion.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's analytical derivations are self-contained and do not reduce to their inputs.

full rationale

The paper derives all central results from stated assumptions using explicit closed-form calculations, with no fitted parameters, no empirical tuning, and no load-bearing self-citations. The equivalence of the two Knill circuits (eqs. (14)-(18) and appendix A), the reduction of Steane correction to a special case of Knill (eqs. (21)-(24) and appendix A), and the Gaussian-spike width formulas (eqs. (33), (47), (49)) are all obtained by direct operator algebra and Gaussian integration, with the key integral proved in lemma 2. The qunaught-based comparison in Section 5 uses the explicitly stated assumption that all resource states have the same squeezing level, and the claimed advantages (symmetric output widths, absence of input-correlated displacements) follow from the derived formulas rather than being inserted by construction. The skeptical concern about eq. (50) and the product metric is a correctness or metric-choice criticism, not a circularity: the paper does not define 'superior' solely through the Heisenberg product, and the equal-resource premise is an explicit assumption rather than a disguised reuse of the conclusion. The only author-provided resource, code in ref. [23], is used for numerical sanity checks and figure generation, not as evidence for any analytical claim. No self-citation chain is invoked to forbid alternatives or to establish the uniqueness of the approach. Therefore the derivation chain is self-contained and the circularity score is 0.

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

No free parameters are fitted; the analysis uses the standard finite-energy parameter Δ and input-to-ancilla width ratios as scenario variables. The assumptions are the common Gaussian-spike approximation, the photon-number dampening noise model, and the equal-Δ resource comparison. No new physical entities are introduced.

assumptions (5)
  • domain assumption Input and ancilla states are well-approximated by linear superpositions of Gaussian spikes sharing a common covariance matrix (Section 4, eqs. (28)-(29)).
    This is the modeling assumption on which the closed-form output width σ²_out (eq. (33)) rests; deviations from equal-width Gaussian spikes are not captured.
  • domain assumption Finite-energy GKP states are described by photon-number dampening e^{-Δ² N}, with peak width Δ and Q ≈ Δ^{-2}I (Section 5, eq. (36)).
    Standard GKP noise model; the paper cites [7] for this and [21] for equivalence of noise models.
  • standard math The beam splitter commutes with the photon-number operator, and the CX gate transforms Gaussian precision matrices via the symplectic matrix S (Sections 5.1-5.2).
    Used to derive the precision matrix Q for the qunaught and standard Bell states.
  • domain assumption The error-corrected state is fully characterized by the peak variance and lattice displacement in each quadrature; finite-energy envelope effects are neglected.
    The formulas (40), (47), (49) report only spike widths; logical error rates also depend on the envelope induced by finite energy.
  • domain assumption All resource states are assigned the same squeezing parameter Δ (Section 5).
    The comparison claims superior performance for equal ancilla quality; if qunaught preparation is more costly, the practical conclusion changes.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Performance analysis of GKP error correction." pith.science (2026). https://pith.science/paper/Y42HMPNK

@misc{pith2026250514775,
  author       = {Pith},
  title        = {Pith review of: Performance analysis of GKP error correction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Y42HMPNK}},
  note         = {Machine review of arXiv:2505.14775}
}
read the original abstract

Quantum error correction is essential for achieving fault-tolerant quantum computing. Gottesman-Kitaev-Preskill (GKP) codes are particularly effective at correcting continuous noise, such as Gaussian noise and loss, and can significantly reduce overhead when concatenated with qubit error-correcting codes like surface codes. GKP error correction can be implemented using either a teleportation-based method, known as Knill error correction, or a quantum non-demolition-based approach, known as Steane error correction. In this work, we conduct a comprehensive performance analysis of these established GKP error correction schemes, deriving an analytical expression for the post-correction GKP squeezing and displacement errors. Our results show that there is flexibility in choosing the entangling gate used with the teleportation-based Knill approach. Furthermore, when implemented using the recently introduced qunaught states, the Knill approach not only achieves superior GKP squeezing compared to other variants but is also the simplest to realize experimentally in the optical domain.

Figures

Figures reproduced from arXiv: 2505.14775 by the authors.

Figure 1
Figure 1. Wave functions in the q-quadrature basis ψ(q) = ⟨q|ψ⟩. (Left) The Logical GKP basis states labelled by |0⟩L and |1⟩L. (Right) Physical states approximating the GKP basis state labelled by |0⟩∆ and |1⟩∆ using the photon number dampening approximation | · ⟩∆ = e −∆ˆn | · ⟩L. error correction (see fig. 2), is based on quantum tele￾portation [14] and was originally developed for gen￾eral stabiliser codes [15] before bei… view at source ↗
Figure 2
Figure 2. Circuits for Knill-type error correction based on using either a controlled displacement (left) or a beam splitter (right) for entangling the input with the ancillas. As demonstrated in section 3, both circuits produce identical error-corrected states, as indicated by the equality sign. |Φ +⟩0,∆ † qˆ pˆ |ψ⟩ s1 |0⟩∆ F s2 |0⟩∆ |ψ ′ ⟩ = (Steane) [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. (Left):Circuit implementation of Steane-type error correction. (right): An equivalent Knill-type error cor￾rection scheme, where the Bell state is prepared using the standard qubit-based preparation circuit. As shown in section 3, Steane-type error correction is a special case of Knill-type error correction. discussions and analyses in the subsequent sections. We define the hermitian quadrature operators qˆ and pˆ, … view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Two methods for preparing an approximation to the Bell state |Φ +⟩L represented by (A) and (B) corre￾sponding to circuit 37 and circuit 41, respectively. The approximation | · ⟩∆ = e −∆ˆn | · ⟩L is used in both cases. (A1): The q-quadrature wavefunction ⟨q1, q2|∅, ∅⟩∆ …
Figure 5
Figure 5. Figure 5: Wigner functions of the output states obtained from numerical simulations of the circuits in fig. 3, applied to the same coherent state with identical measurement outcomes. Simulations were performed using the Strawberry Fields Python package [22]. Both visual inspecti…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Impact of finite squeezing on near-term quantum computations using GKP qubits

    quant-ph 2025-07 conditional novelty 7.0 of 10

    A 108-mode simulation of a GKP-based measurement-based quantum computer shows a three-qubit Grover search beats the classical one-query bound only above about 10 dB of GKP squeezing.

Reference graph

Works this paper leans on

24 extracted references · 13 canonical work pages · cited by 1 Pith paper

  1. [20]

    Correcting biased noise using Gottesman-Kitaev-Preskill repetition code with noisy ancilla

    Z. Li and D. Su. Correcting biased noise using Gottesman-Kitaev-Preskill repetition code with noisy ancilla. en. arXiv:2308.01549 [quant-ph]. Aug. 2023

    work page Pith review arXiv 2023

  2. [1]

    Algorithms for quantum computation: discrete logarithms and factoring

    P. Shor. “Algorithms for quantum computation: discrete logarithms and factoring”. In: Proceed- ings 35th Annual Symposium on Foundations of Computer Science. Santa Fe, NM, USA: IEEE Comput. Soc. Press, 1994, pp. 124–134.doi: 10. 1109/SFCS.1994.365700. 9

    arXiv 1994

  3. [2]

    Gottesman

    D. Gottesman. An Introduction to Quantum Error Correction and Fault-Tolerant Quantum Computation. arXiv:0904.2557 [quant-ph]. Apr

    arXiv

  4. [3]

    Encoding a qubit in an os- cillator

    D. Gottesman et al. “Encoding a qubit in an os- cillator”. In:Physical Review A64.1 (June 2001). Publisher: American Physical Society, p. 012310. doi: 10.1103/PhysRevA.64.012310

    doi:10.1103/physreva.64.012310 2001

  5. [4]

    Fault-Tolerant Measurement- Based Quantum Computing with Continuous- Variable Cluster States

    N. C. Menicucci. “Fault-Tolerant Measurement- Based Quantum Computing with Continuous- Variable Cluster States”. en. In:Physical Review Letters 112.12 (Mar. 2014), p. 120504.doi: 10. 1103/PhysRevLett.112.120504

    work page 2014

  6. [5]

    Fault-Tolerant Continuous- Variable Measurement-based Quantum Compu- tation Architecture

    M. V. Larsen et al. “Fault-Tolerant Continuous- Variable Measurement-based Quantum Compu- tation Architecture”. en. In: PRX Quantum 2.3 (Aug. 2021), p. 030325. doi: 10 . 1103 / PRXQuantum.2.030325

    work page 2021

  7. [6]

    Fault-Tolerant Quantum Com- putation with Static Linear Optics

    I. Tzitrin et al. “Fault-Tolerant Quantum Com- putation with Static Linear Optics”. In: PRX Quantum 2.4 (Dec. 2021). Publisher: American Physical Society, p. 040353. doi: 10 . 1103 / PRXQuantum.2.040353

    work page 2021

  8. [7]

    Fault-tolerant bosonic quantum error correction with the surface–Gottesman-Kitaev-Preskill code

    K. Noh and C. Chamberland. “Fault-tolerant bosonic quantum error correction with the surface–Gottesman-Kitaev-Preskill code”. In: Physical Review A 101.1 (Jan. 2020). Pub- lisher: American Physical Society, p. 012316. doi: 10.1103/PhysRevA.101.012316

    doi:10.1103/physreva.101.012316 2020

Show all 24 references
  1. [8]

    Analog Quantum Error Correc- tion with Encoding a Qubit into an Oscillator

    K. Fukui et al. “Analog Quantum Error Correc- tion with Encoding a Qubit into an Oscillator”. In: Physical Review Letters119.18 (Nov. 2017). Publisher: American Physical Society, p. 180507. doi: 10.1103/PhysRevLett.119.180507

    2017 doi

  2. [9]

    Low-Overhead Fault-Tolerant Quantum Error Correction with the Surface- GKP Code

    K. Noh et al. “Low-Overhead Fault-Tolerant Quantum Error Correction with the Surface- GKP Code”. en. In: PRX Quantum 3.1 (Jan. 2022), p. 010315. doi: 10 . 1103 / PRXQuantum . 3.010315

    2022

  3. [10]

    Real-time quantum error correction beyond break-even

    V. V. Sivak et al. “Real-time quantum error correction beyond break-even”. en. In: Nature 616.7955 (Apr. 2023). Publisher: Nature Publish- ing Group, pp. 50–55. doi: 10 . 1038 / s41586 - 023-05782-6

    2023

  4. [11]

    Beating the break-even point with a discrete-variable-encoded logical qubit

    Z. Ni et al. “Beating the break-even point with a discrete-variable-encoded logical qubit”. en. In: Nature 616.7955 (Apr. 2023). Publisher: Nature Publishing Group, pp. 56–60. doi: 10 . 1038 / s41586-023-05784-4

    2023

  5. [12]

    Error correction of a logical grid state qubit by dissipative pumping

    B. de Neeve et al. “Error correction of a logical grid state qubit by dissipative pumping”. en. In: Nature Physics18.3 (Mar. 2022). Publisher: Na- ture Publishing Group, pp. 296–300. doi: 10 . 1038/s41567-021-01487-7

    2022

  6. [13]

    Logical states for fault-tolerant quantum computation with propagating light

    S. Konno et al. “Logical states for fault-tolerant quantum computation with propagating light”. In: Science 383.6680 (Jan. 2024). Publisher: American Association for the Advancement of Science, pp. 289–293. doi: 10 . 1126 / science . adk7560

    2024

  7. [14]

    Demonstrating the viability of universal quantum computation using teleportation and single-qubit operations

    D. Gottesman and I. L. Chuang. “Demonstrating the viability of universal quantum computation using teleportation and single-qubit operations”. en. In: Nature 402.6760 (Nov. 1999). Num- ber: 6760 Publisher: Nature Publishing Group, pp. 390–393. doi: 10.1038/46503

    1999 doi

  8. [15]

    Scalable quantum computing in the presence of large detected-error rates

    E. Knill. “Scalable quantum computing in the presence of large detected-error rates”. en. In: Physical Review A71.4 (Apr. 2005), p. 042322. doi: 10.1103/PhysRevA.71.042322

    2005 doi

  9. [16]

    Progress towards practical qubit computation using approximate Gottesman- Kitaev-Preskill codes

    I. Tzitrin et al. “Progress towards practical qubit computation using approximate Gottesman- Kitaev-Preskill codes”. en. In: Physical Review A 101.3 (Mar. 2020), p. 032315.doi: 10.1103/ PhysRevA.101.032315

    2020

  10. [17]

    Continuous-variable gate teleportation and bosonic-code error correction

    B. W. Walshe et al. “Continuous-variable gate teleportation and bosonic-code error correction”. en. In: Physical Review A 102.6 (Dec. 2020), p.062411. doi: 10.1103/PhysRevA.102.062411

    2020 doi

  11. [18]

    Error Correcting Codes in Quan- tum Theory

    A. M. Steane. “Error Correcting Codes in Quan- tum Theory”. In: Physical Review Letters 77.5 (July 1996). Publisher: American Physical Soci- ety, pp. 793–797. doi: 10.1103/PhysRevLett. 77.793

    1996 doi

  12. [19]

    All-Gaussian Universality and Fault Tolerance with the Gottesman-Kitaev- Preskill Code

    B. Q. Baragiola et al. “All-Gaussian Universality and Fault Tolerance with the Gottesman-Kitaev- Preskill Code”. en. In: Physical Review Letters 123.20 (Nov. 2019), p. 200502. doi: 10 . 1103 / PhysRevLett.123.200502

    2019

  13. [21]

    Equivalence of approx- imate Gottesman-Kitaev-Preskill codes

    T. Matsuura et al. “Equivalence of approx- imate Gottesman-Kitaev-Preskill codes”. en. In: Physical Review A 102.3 (Sept. 2020). arXiv:1910.08301 [quant-ph], p. 032408.doi: 10. 1103/PhysRevA.102.032408

    2020 arXiv

  14. [22]

    Strawberry Fields: A Software Platform for Photonic Quantum Computing

    N. Killoran et al. “Strawberry Fields: A Software Platform for Photonic Quantum Computing”. en-GB. In: Quantum 3 (Mar. 2019). Publisher: Verein zur Förderung des Open Access Pub- lizierens in den Quantenwissenschaften, p. 129. doi: 10.22331/q-2019-03-11-129

    2019 doi

  15. [23]

    F. K. Marqversen. quantum_computations. https : / / github . com / frederik - kofoed - marqversen/quantum_computations. 10 Appendices 11 A Explicit calculations for GKP error correction circuits We explicitly calculate the action of the different types of GKP error correctionK...

  16. [2009]

    doi: 10.48550/arXiv.0904.2557

    doi arXiv

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.