Pith. sign in

REVIEW 3 major objections 6 minor 29 references

Faster Probabilistic Error Cancellation

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 This paper proposes a faster variant of probabilistic error cancellation that expands each inverse error channel into identity plus an error generator, regroups the circuit by powers of that generator, and allocates shots…

desk verdict A genuinely cheaper PEC variant, but the cost saving and unbiasedness depend on a per-gate uniform noise assumption that the paper underplays. read the letter →

arxiv 2506.04468 v1 pith:TKUQL7NQ submitted 2025-06-04 quant-ph

Yi-Hsiang Chen This is my paper
classification quant-ph
keywords probabilisticerrorcancellationquantummitigationsamplingoverheadtruncatedseriesexpansiondeterministicshotallocationzero-noiseextrapolationstochasticPaulichannelTrottersimulation
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 introduces Faster Probabilistic Error Cancellation (FPEC), a protocol for quantum error mitigation that recovers the noiseless expectation value as a weighted combination of noisy circuits. Instead of sampling each gate's inverse channel stochastically as standard PEC does, FPEC decomposes each inverse channel into an identity part and an error generator, expands the whole circuit in powers of that generator, and drops the smallest terms. It claims this is more cost-effective than standard PEC: series truncation lowers the overhead, and deterministic shot allocation removes an extra variance term that appears when circuits are sampled randomly. This gives a user-controlled trade-off between bias and statistical noise, and in numerical tests on a two-dimensional transverse-field Ising model FPEC needed about half as many shots as standard PEC for the same statistical error. The paper also reports hardware runs where mitigated values matched exact values, and cases where FPEC beats zero-noise extrapolation because its bias is controlled.

What carries the argument

The central object is the binomial expansion of the inverse error channel. For an error channel $\Lambda = (1-\epsilon)I + \epsilon E'$, write $\Lambda^{-1} = (1+\epsilon_1)I - \epsilon_2 E$, where $E$ is the 'inverse generator,' a normalized combination of implementable unitaries such as Pauli operators. Expanding the $l$-fold product of inverse channels and grouping terms by how many $E$'s appear turns the ideal circuit into a weighted sum over powers of $E$; this is what makes truncation natural and lets shots be assigned deterministically to each $C_k$.

What would settle it

A direct test is to run FPEC on a circuit with known identical depolarizing channels per gate and compare the realized estimator variance with Eq. (7): if it exceeds standard PEC's variance for the same shot count, the claimed saving is not there. A sharper test is to introduce heterogeneous gate error rates (for example alternating $10^{-4}$ and $10^{-2}$) while the protocol still uses a single $\Lambda$; if the achieved bias or variance is no better than standard PEC, the identical-channel assumption is essential.

Watch

Extended reading notes

Core claim

In the paper's own framing, the ideal circuit is exactly recovered as a linear combination of noisy circuits: $C_{\text{ideal}} = \sum_{k=0}^{l} \binom{l}{k}(1+\epsilon_1)^{l-k}(-\epsilon_2)^k C_k$, where $C_k$ is the average over circuits with $k$ insertions of the inverse generator $E$. The paper's estimator keeps only the first $K$ terms, with shots distributed across those terms in proportion to their coefficient magnitudes. This is claimed to be more efficient than standard PEC for two reasons: truncation shrinks the overhead from $(1+2\epsilon)^{2l}$ to $\left(\sum_{k=0}^{K}|\gamma_k|\right)^2$, and deterministic allocation removes a nonnegative variance gap $\Delta$ that appears when circuits are instead sampled with probabilities $|\gamma_k|/\sum_j |\gamma_j|$. The truncation bias is controlled by the bound $\|O\|\sum_{k=K+1}^{l}|\gamma_k|$, so the user can choose $K$ to make residual bias smaller than the statistical noise.

Load-bearing premise

The binomial regrouping in Eq. (2) assumes every gate or layer shares the same error channel $\Lambda$, so that different $k$-error subsets collapse into a single term with weight $\gamma_k$; heterogeneous gate errors make the coefficients position-dependent and the deterministic-allocation saving no longer follows.

Editorial extensions

If this is right

  • With the same number of shots, FPEC yields a mitigated expectation value with lower variance than standard PEC whenever the truncated series omits negligible weights and the noisy observables $\langle O\rangle_k$ differ from one another.
  • A user can set a bias tolerance $\delta$, choose the smallest $K$ with $\|O\|\sum_{k=K+1}^{l}|\gamma_k| \leq \delta$, and thereby guarantee that the residual bias is below statistical relevance.
  • In the global-depolarizing limit, the $K=0$ estimator $(1+\epsilon_1)^l\langle O\rangle_0$ saturates the fundamental sampling-cost lower bound, making FPEC optimal in that case.
  • In numerical 2D transverse-field Ising simulations, FPEC required about half the shots of standard PEC to reach the same statistical error, and in comparisons with ZNE the controlled bias let FPEC win when ZNE's ansatz bias dominated.

Reading between the lines

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

  • Because the deterministic-allocation saving grows with the spread among the $\langle O\rangle_k$, circuits whose observables are strongly affected by where errors occur will benefit most; one could pre-compute or estimate this spread to decide when FPEC is worthwhile.
  • The identical-channel assumption is likely the main practical obstacle: a natural extension would track position-dependent coefficients for heterogeneous gate errors, at the cost of heavier bookkeeping, and the paper does not analyze whether the savings survive in that setting.
  • FPEC's truncation parameter $K$ interpolates between a cheap biased estimate (small $K$, ZNE-like) and the full unbiased PEC, suggesting that $K$ could be tuned online based on measured shot noise rather than fixed in advance.
  • The experimental robustness to a depolarizing-channel mischaracterization hints that the protocol may tolerate approximate error models, but this is only demonstrated for one gate set and should be tested on structured non-Pauli noise.
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 proposes Faster PEC (FPEC), a variant of probabilistic error cancellation. Starting from an inverse error channel written as identity plus a correction, Λ^{-1} = (1+ε1)I - ε2E, the protocol expands a depth-l noisy circuit as a binomial sum over powers of the inverse generator E, Eq. (2). It then estimates the grouped noisy circuits C_k, truncates the series at an order K selected either by a user bias tolerance δ or by the available shot budget M, allocates M|γ_k|/∑|γ_j| shots to each term, and forms the estimator ∑_{k=0}^K γ_k ⟨O⟩_k. The paper derives the per-shot variance in Eq. (7), proves in Eq. (8) that deterministic shot allocation dominates a sampling-based allocation, and supports the efficiency claim with a 2D transverse-field Ising model simulation and a 20-qubit Quantinuum H1 experiment.

Significance. The deterministic allocation of shots among the grouped circuits is a clean and generally valid variance-reduction step, and the explicit bias-versus-shots trade-off through truncation is practically appealing. The analytic formulas are mostly transparent, and the experimental demonstration on H1 is a strength. The main limitation is that the central binomial expansion is derived under an identical-per-gate error channel assumption; without that assumption the protocol as written does not implement the inverse channel, so the general efficiency claim in the conclusion is broader than what is proven.

major comments (3)
  1. [Section 3, Eq. (2)] The expansion C_ideal = ∑_{k} γ_k C_k is only valid when every gate (or circuit layer) shares the same error channel Λ. If gates have position-dependent channels Λ_i = (1-ε_i)I + ε_i E_i, then the coefficient of a subset S is ∏_{i∉S}(1+ε_{1i})∏_{i∈S}(-ε_{2i}), which depends on the positions in S. In that case Eq. (2) and the uniform sampling of k locations in Eq. (5) do not reconstruct the ideal circuit even in the infinite-shot limit, because the grouped coefficient γ_k is not well defined. The claim in Sec. 2 that the method works 'without extra assumptions on the error channel or the circuit structure' is therefore unsupported. The numerical simulation in Sec. 4 and the experiment in Sec. 5 both attach a single measured channel to every two-qubit gate, so neither test covers heterogeneous per-gate noise. Please either provide the position-dependent generalization (group subsets by k with weights a_S = ∏_{i∉S}(1+ε_{1i})∏_{i∈S}(-ε_{2i}) and sample locations from the normalized absolute weights) or explicitly restrict the stated claims to circuits with identical per-gate or per-layer error channels.
  2. [Section 3.1, Eq. (7)] The analytic saving from truncation is derived under the assumption that Var[O_k] is similar for all k. The paper provides no general bound on how Var[O_k] varies with k, so the conclusion in Sec. 6 that FPEC is 'more efficient than the standard PEC in terms of the total sampling cost' is not established as a general theorem. What is proven is a variance formula for FPEC and a rigorous variance reduction relative to a sampling-based allocation over k; the overall efficiency advantage over standard PEC is demonstrated in a specific numerical example. Please state the efficiency claim as conditional on the variance profile of the grouped circuits or supply a worst-case bound that justifies the unconditional wording.
  3. [Section 5] The experimental demonstration infers a single depolarizing channel for the non-Clifford two-qubit gate and applies it to all gates; the footnote correctly acknowledges possible mischaracterization, but the robustness of the protocol to heterogeneous or mischaracterized channels is not quantified. Given that the uniform-channel assumption is load-bearing for Eq. (2), the experimental agreement alone does not validate the method under realistic gate-dependent noise, and this limitation should be discussed explicitly in the conclusion.
minor comments (6)
  1. [Section 4] The text says 'We ran total M = 5000 shots and computed the standard deviation σ_PEC of the 5000 values and deduce the variance ... = σ^2_PEC.' This defines a per-shot variance; the standard error of the mean would be σ_PEC/√M. Please clarify which quantity is plotted in Fig. 2 and how the estimates are obtained reproducibly.
  2. [Figure 3] No error bars or confidence intervals are shown for either method, so the claim that the FPEC bias 'is due to shot noise' is not visually supported; please add error bars or report the relevant statistical uncertainties.
  3. [Section 3, protocol step 4] The instruction 'sampling k locations uniformly at random from total l positions' is ambiguous when E can only be inserted at certain gate types (for example, only two-qubit gates); please specify the exact set of available locations and how l is defined in that case.
  4. [Section 3, after Eq. (5)] The statement 'Hoeffding's inequality then guarantees a fast convergence' is loose: Hoeffding gives a standard O(1/√n) concentration bound for bounded observables, not a special speed-up. Please rephrase to avoid overclaiming.
  5. [General] There are several minor grammatical issues; for example, 'the number of locations binom(l,k) can be very large that directly averaging over those circuit values requires too many circuits to be run' should read 'so large that directly averaging ... requires too many circuits to be run.'
  6. [Eq. (8)] The notation p(b|k) and ⟨b|O|b⟩_k is introduced but the subscript k on the bra-ket is not defined; please define what is meant by the k-th circuit's bitstring expectation value.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the FPEC saving follows algebraically from the stated identical-channel expansion and an explicitly chosen shot allocation, with independent experimental inputs.

full rationale

The derivation chain is self-contained. Eq. (2) follows algebraically from the stated assumption that every gate carries the same channel Λ = (1−ϵ)I + ϵE′ and from the inverse form in Eq. (1); no fitted parameter is later renamed as a prediction. The truncation order K is selected from |γ_k|, a bias tolerance δ, or the shot count M, and the deterministic allocation M|γ_k|/Σ|γ_j| is a user choice, not a parameter fit to the target observable. The variance comparison in Eqs. (7)–(8) is a mathematical identity for that allocation plus a Cauchy–Schwarz inequality against probabilistic sampling, so the reported saving is a property of the protocol definition rather than a circular prediction. The experimental mitigation uses an independently measured randomized-benchmarking infidelity, and the ideal values are compared externally. The only notable gap is correctness/robustness, not circularity: the identical-channel assumption in Sec. 3 conflicts with the Sec. 2 sentence claiming the method works 'without extra assumptions on the error channel or the circuit structure,' and heterogeneous per-gate channels would break the single-γ_k grouping. That is an unsupported assumption, not a reduction of the result to its inputs.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The protocol relies on standard PEC-style decomposability and a uniform per-gate noise model. The only user-chosen numbers are the truncation order K and bias tolerance delta. No new physical entities are introduced.

free parameters (2)
  • Truncation order K = 1-4 in the H1 experiment; determined by bias tolerance (0.01) or shot count
    K sets the number of terms retained in the binomial expansion; it is a user-chosen accuracy/variance trade-off, not fitted to data, but it directly determines the saving claimed.
  • Bias tolerance delta = 0.001 (simulation, Fig. 3), 0.01 (experiment)
    User-set bound for the truncation bias; controls how aggressively the series is truncated.
assumptions (4)
  • domain assumption The circuit consists of l gates, each subject to the same error channel Lambda = (1 - epsilon)I + epsilon E'.
    Section 3 defines C = Lambda U_l ... Lambda U_1 with a single Lambda; this permits grouping terms by binomial coefficients. Real devices have heterogeneous gate errors.
  • domain assumption The inverse channel has the form Lambda^-1 = (1 + epsilon_1)I - epsilon_2 E with E = sum_i c_i V_i(.)V_i^dagger and sum_i |c_i| = 1.
    Section 3, Eq. (1); standard PEC decomposability, valid for stochastic Pauli channels.
  • standard math The observable O is bounded with known operator norm ||O|| (normalized observable ||O|| = 1).
    Used for the truncation bias bound in Section 3: |sum_{k>K} gamma_k <O>_k| <= ||O|| sum_{k>K} |gamma_k|.
  • standard math Uniform sampling of positions and unitaries yields a Hoeffding-convergent estimate of <O>_k.
    Section 3 invokes Hoeffding's inequality for the Monte Carlo estimate of <O>_k over (l choose k) positions.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Faster Probabilistic Error Cancellation." pith.science (2026). https://pith.science/paper/TKUQL7NQ

@misc{pith2026250604468,
  author       = {Pith},
  title        = {Pith review of: Faster Probabilistic Error Cancellation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TKUQL7NQ}},
  note         = {Machine review of arXiv:2506.04468}
}
read the original abstract

Probabilistic error cancellation (PEC) is a leading quantum error mitigation method that provides an unbiased estimate, although it is known to have a large sampling overhead. In this work, we propose a new method to perform PEC, which results in a lower sampling cost than the standard way. It works by decomposing the inverse channel of each gate or each circuit layer into the identity part and the non-identity part and reorganizing the full circuit as different powers of the inverse generator. The ideal circuit becomes a linear combination of noisy circuits with different weights where shots are deterministically allocated to each circuit based on its weight. This naturally sets the achievable bias given a finite amount of shots. As the number of shots is increased, smaller bias terms can be gradually resolved and become bias-free in the limit of sufficient shots. We show the saving both analytically and numerically over the standard PEC and identify situations where it can outperform heuristic approach, such as zero-noise extrapolation, due to the well-controlled bias. We also demonstrated this method experimentally and found excellent agreement between the mitigated and the ideal values.

Figures

Figures reproduced from arXiv: 2506.04468 by the authors.

Figure 1
Figure 1. The coefficients |γk| as a function k, with l = 1000 total number of gates and error rate ϵ = 10−3 under a depolarizing error channel Λ. As shown in [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Comparison of the variance of the raw (unmitigated) value and the mitigated values from [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. The comparison of the absolute observable error in the mitigated value between FPEC and [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Left: The expectation value ⟨S 2 z ⟩ := ⟨( PN j=1 Zj/N) 2 ⟩ as a function of the number of Trotter steps. Right: The expectation value ⟨Zavg⟩ := ⟨ PN j=1 Qj i=1 Zi  /N⟩ as a function of the number of Trotter steps. “H1 faster PEC” represents FPEC protocol and “H1 raw…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

29 extracted references · 9 canonical work pages

  1. [1]

    Lindner, Ron Aharon Melcer, Adiel Meyer, Gili Schul, and Maor Shutman

    Dorit Aharonov, Ori Alberton, Itai Arad, Yosi Atia, Eyal Bairey, Zvika Brakerski, Itsik Cohen, Omri Golan, Ilya Gurwich, Oded Kenneth, Eyal Leviatan, Netanel H. Lindner, Ron Aharon Melcer, Adiel Meyer, Gili Schul, and Maor Shutman. On the importanceoferrormitigationforquantumcomputation, 2025. URL https://arxiv. org/abs/2503.17243

    arXiv 2025

  2. [2]

    Multi-exponential error extrapolation and combining error mit- igation techniques for nisq applications

    Zhenyu Cai. Multi-exponential error extrapolation and combining error mit- igation techniques for nisq applications. npj Quantum Information , 7(1):80,

    work page

  3. [3]

    Benjamin, Suguru Endo, William J

    Zhenyu Cai, Ryan Babbush, Simon C. Benjamin, Suguru Endo, William J. Huggins, Ying Li, Jarrod R. McClean, and Thomas E. O’Brien. Quantum error mitigation. Rev. Mod. Phys. , 95:045005, Dec 2023. DOI: 10.1103/RevModPhys.95.045005. URL https://link.aps.org/doi/10.1103/RevModPhys.95.045005

    doi:10.1103/revmodphys.95.045005 2023

  4. [4]

    Tran, and Patrick Rall

    Andrew Eddins, Minh C. Tran, and Patrick Rall. Lightcone shading for classically accelerated quantum error mitigation, 2024. URL https://arxiv.org/abs/2409. 04401

    work page 2024

  5. [5]

    Tudor Giurgica-Tiron, Yousef Hindy, Ryan LaRose, Andrea Mari, and William J. Zeng. Digital zero noise extrapolation for quantum error mitigation. In2020 IEEE International Conference on Quantum Computing and Engineering (QCE) , pages 306– 316, 2020. DOI: 10.1109/QCE49297.2020.00045

    arXiv 2020

  6. [6]

    Dilution of error in digital hamiltonian simulation

    Etienne Granet and Henrik Dreyer. Dilution of error in digital hamiltonian simulation. PRX Quantum , 6:010333, Feb 2025. DOI: 10.1103/PRXQuantum.6.010333. URL https://link.aps.org/doi/10.1103/PRXQuantum.6.010333

    work page doi:10.1103/prxquantum.6.010333 2025

  7. [7]

    Introduction to quantum gate set tomography, 2015

    Daniel Greenbaum. Introduction to quantum gate set tomography, 2015. URLhttps: //arxiv.org/abs/1509.02921

    arXiv 2015

  8. [8]

    Noise effects on purity and quantum entangle- ment in terms of physical implementability

    Yuchen Guo and Shuo Yang. Noise effects on purity and quantum entangle- ment in terms of physical implementability. npj Quantum Information , 9(1): 11, 2023. DOI: 10.1038/s41534-023-00680-1. URL https://doi.org/10.1038/ s41534-023-00680-1

    doi:10.1038/s41534-023-00680-1 2023

Show all 29 references
  1. [9]

    Rudolph, CristinaCirstoiu, KevinHemery, ConorMcKeever, MichaelLubasch, EtienneGranet, Charles H

    Reza Haghshenas, Eli Chertkov, Michael Mills, Wilhelm Kadow, Sheng-Hsuan Lin, Yi-Hsiang Chen, Chris Cade, Ido Niesen, Tomislav Begušić, Manuel S. Rudolph, CristinaCirstoiu, KevinHemery, ConorMcKeever, MichaelLubasch, EtienneGranet, Charles H. Baldwin, John P. Bartolotta, Matth...

    2025 arXiv

  2. [10]

    Physical Implementability of Linear Maps and Its Application in Error Mitigation

    Jiaqing Jiang, Kun Wang, and Xin Wang. Physical Implementability of Linear Maps and Its Application in Error Mitigation. Quantum, 5:600, December 2021. ISSN 2521-327X. DOI: 10.22331/q-2021-12-07-600. URL https://doi.org/10.22331/ q-2021-12-07-600

    2021 doi

  3. [11]

    Evidence for the utility of quantum computing before fault tolerance

    Youngseok Kim, Andrew Eddins, Sajant Anand, Ken Xuan Wei, Ewout van den Berg, Sami Rosenblatt, Hasan Nayfeh, Yantao Wu, Michael Zaletel, Kristan Temme, and Abhinav Kandala. Evidence for the utility of quantum computing before fault tolerance. Nature, 618(7965):500–505, 2023. D...

    2023 doi

  4. [12]

    Theory of quantum error mitiga- tion for non-clifford gates, 2025

    David Layden, Bradley Mitchell, and Karthik Siva. Theory of quantum error mitiga- tion for non-clifford gates, 2025. URLhttps://arxiv.org/abs/2403.18793

    2025

  5. [14]

    Andrea Mari, Nathan Shammah, and William J. Zeng. Extending quantum prob- abilistic error cancellation by noise scaling. Phys. Rev. A , 104:052607, Nov 2021. DOI: 10.1103/PhysRevA.104.052607. URL https://link.aps.org/doi/10.1103/ PhysRevA.104.052607

    2021 doi

  6. [15]

    Stemen, Misty Wahl, William J

    Benjamin McDonough, Andrea Mari, Nathan Shammah, Nathaniel T. Stemen, Misty Wahl, William J. Zeng, and Peter P. Orth. Automated quantum error mitiga- tion based on probabilistic error reduction. In 2022 IEEE/ACM Third Interna- tional Workshop on Quantum Computing Software (QCS...

    2022

  7. [16]

    S. A. Moses, C. H. Baldwin, M. S. Allman, R. Ancona, L. Ascarrunz, C. Barnes, J. Bartolotta, B. Bjork, P. Blanchard, M. Bohn, J. G. Bohnet, N. C. Brown, N. Q. Burdick, W. C. Burton, S. L. Campbell, J. P. Campora, C. Carron, J. Chambers, J. W. Chan, Y. H. Chen, A. Chernoguzov, ...

    2023

  8. [17]

    Quasiprobability de- compositions with reduced sampling overhead

    Christophe Piveteau, David Sutter, and Stefan Woerner. Quasiprobability de- compositions with reduced sampling overhead. npj Quantum Information , 8(1): 12, 2022. DOI: 10.1038/s41534-022-00517-3. URL https://doi.org/10.1038/ s41534-022-00517-3

    2022 doi

  9. [18]

    Proctor, Arnaud Carignan-Dugas, Kenneth Rudinger, Erik Nielsen, Robin 11 Blume-Kohout, andKevinYoung

    Timothy J. Proctor, Arnaud Carignan-Dugas, Kenneth Rudinger, Erik Nielsen, Robin 11 Blume-Kohout, andKevinYoung. Directrandomizedbenchmarkingformultiqubitde- vices. Phys. Rev. Lett., 123:030503, Jul 2019. DOI: 10.1103/PhysRevLett.123.030503. URL https://link.aps.org/doi/10.110...

    2019 doi

  10. [19]

    Operational applications of the diamond normandrelatedmeasuresinquantifyingthenon-physicalityofquantummaps

    Bartosz Regula, Ryuji Takagi, and Mile Gu. Operational applications of the diamond normandrelatedmeasuresinquantifyingthenon-physicalityofquantummaps. Quan- tum, 5:522, August 2021. ISSN 2521-327X. DOI: 10.22331/q-2021-08-09-522. URL https://doi.org/10.22331/q-2021-08-09-522

    2021 doi

  11. [20]

    Reducing pec overhead by pauli error propagation, 2025

    Timon Scheiber, Paul Haubenwallner, and Matthias Heller. Reducing pec overhead by pauli error propagation, 2025. URLhttps://arxiv.org/abs/2412.01311

    2025 arXiv

  12. [21]

    Quantum error mitigation as a universal error reduction technique: Applications from the nisq to the fault-tolerant quantum computing eras

    Yasunari Suzuki, Suguru Endo, Keisuke Fujii, and Yuuki Tokunaga. Quantum error mitigation as a universal error reduction technique: Applications from the nisq to the fault-tolerant quantum computing eras. PRX Quantum , 3:010345, Mar 2022. DOI: 10.1103/PRXQuantum.3.010345. URLh...

    2022 doi

  13. [22]

    Optimal resource cost for error mitigation.Phys

    Ryuji Takagi. Optimal resource cost for error mitigation.Phys. Rev. Res., 3:033178, Aug 2021. DOI: 10.1103/PhysRevResearch.3.033178. URLhttps://link.aps.org/ doi/10.1103/PhysRevResearch.3.033178

    2021 doi

  14. [23]

    Gambetta

    Kristan Temme, Sergey Bravyi, and Jay M. Gambetta. Error mitigation for short- depth quantum circuits.Phys. Rev. Lett., 119:180509, Nov 2017. DOI: 10.1103/Phys- RevLett.119.180509. URL https://link.aps.org/doi/10.1103/PhysRevLett. 119.180509

    2017 doi

  15. [24]

    Tran, Kunal Sharma, and Kristan Temme

    Minh C. Tran, Kunal Sharma, and Kristan Temme. Locality and error mitigation of quantum circuits, 2023. URLhttps://arxiv.org/abs/2303.06496

    2023 arXiv

  16. [25]

    Universal cost bound of quantum error mitigation based on quantum estimation theory.Phys

    Kento Tsubouchi, Takahiro Sagawa, and Nobuyuki Yoshioka. Universal cost bound of quantum error mitigation based on quantum estimation theory.Phys. Rev. Lett. , 131:210601, Nov 2023. DOI: 10.1103/PhysRevLett.131.210601. URLhttps://link. aps.org/doi/10.1103/PhysRevLett.131.210601

    2023 doi

  17. [26]

    Sym- metric clifford twirling for cost-optimal quantum error mitigation in early ftqc regime,

    Kento Tsubouchi, Yosuke Mitsuhashi, Kunal Sharma, and Nobuyuki Yoshioka. Sym- metric clifford twirling for cost-optimal quantum error mitigation in early ftqc regime,

  18. [27]

    Zoltán Zimborás, Bálint Koczor, Zoë Holmes, Elsi-Mari Borrelli, András Gilyén, Hsin- Yuan Huang, Zhenyu Cai, Antonio Acín, Leandro Aolita, Leonardo Banchi, Fernando G. S. L. Brandão, Daniel Cavalcanti, Toby Cubitt, Sergey N. Filippov, Guillermo García-Pérez, John Goold, Orsoly...

  19. [30]

    URL https://arxiv.org/abs/2501.05694. 12

    arXiv

  20. [2021]

    URL https://doi.org/10.1038/ s41534-021-00404-3

    DOI: 10.1038/s41534-021-00404-3. URL https://doi.org/10.1038/ s41534-021-00404-3

    doi

  21. [2025]

    URL https://arxiv.org/abs/2405.07720

    arXiv

Pith tools

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