Pith. sign in

REVIEW 2 major objections 4 minor 79 references

Optimal strategies for shadow tomography with limited resources

T0 review · 2 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Shadow-tomographing Pauli observables optimally reduces to a single graph number, and Clifford measurements achieve it for all one- and two-qubit blocks.

desk verdict Solid extension of the beta-number framework to block-local shadow tomography; the two-qubit Clifford optimality rests on a companion-paper classification that should be visible to referees. read the letter →

arxiv 2608.08429 v1 pith:3VOU74CU submitted 2026-08-09 quant-ph

Zhen-Peng Xu , Qi Ye , Jie Wang , Gereon Ko{ss}mann , Andreas Winter , René Schwonnek This is my paper · ORCID
classification quant-ph MSC 81P6805C1505C6990C22
keywords shadowtomographyPauliobservablesfrustrationgraphsamplecomplexityCliffordmeasurementshbar-perfectgraphsfractionalchromaticnumberHamiltonianenergyestimation
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

Shadow tomography promises to estimate many expectation values of an unknown quantum state from few copies, but existing guarantees rarely come with strategies that can be implemented on real devices. This paper shows that, for Pauli observables measured one copy at a time with measurements restricted to fixed blocks of qubits, the optimal sample-complexity parameter is controlled by a single graph built from Pauli anticommutation relations. The central result pins this parameter between the reciprocal fractional chromatic number and the reciprocal Lovász number of the union frustration graph; when the local graphs are $\hbar$-perfect, or the union graph is perfect, the lower bound is exact and Clifford measurements achieve it. This makes optimal shadow tomography constructive for all one-qubit and two-qubit measurement blocks, and it yields concrete variance bounds for molecular Hamiltonian energy estimation.

What carries the argument

The load-bearing object is the frustration graph of a set of Pauli observables: vertices are observables, edges join anticommuting pairs; for a block partition the paper forms the edge-union $G(\tau)$ of the frustration graphs of the local restrictions. The key identity is $\delta_\tau(\{S_i\}) = \min_{w\in\Delta}\max_{\rho\in \mathrm{PD}_\tau} \sum_i w_i \langle S_i\rangle_\rho^2$, which reduces sample complexity to an optimization over product states. The comparison theorem then sandwiches this quantity between $1/\chi_f(G(\tau))$ and $1/\vartheta(\overline{G(\tau)})$. The class of $\hbar$-perfect graphs—graphs for which the weighted $\beta$ number $\beta(G,w)=\max_\rho\sum_iw_i\langle S_i\rangle_\rho^2$ equals the weighted independence number $\alpha(G,w)$ for all $w$—is exactly what makes the lower bound exact, because on such graphs Clifford measurements attain the fractional chromatic number.

What would settle it

Enumerate all two-qubit Pauli realizations, compute for each resulting frustration graph the weighted $\beta$ number via SDP and compare with the weighted independence number; finding any two-qubit graph with $\beta(G,w)>\alpha(G,w)$ for some weight vector $w$ would make the claimed Clifford optimality for two-qubit partitions false. Alternatively, on a concrete two-qubit partition compute $\delta_\tau$ by the SDP hierarchy and compare with $1/\chi_f(G(\tau))$; a gap would refute Corollary 4.

Watch

Extended reading notes

Core claim

For a set of Pauli observables and a partition $\tau$ of the qubits into independently measured blocks, the paper defines an optimal sample-complexity parameter $\delta_\tau(\{S_i\})$ and proves $\delta_\tau(\{S_i\})\in[1/\chi_f(G(\tau)),1/\vartheta(\overline{G(\tau)})]$, where $G(\tau)$ is the edge-union of the local frustration graphs of the observables' restrictions to each block. The lower bound is tight—and achievable by randomizing Clifford measurements—whenever each local frustration graph is $\hbar$-perfect, meaning its $\beta$ number coincides with its weighted independence number for every weight vector, or when $G(\tau)$ is a perfect graph. Consequently, for any partition into pairs of qubits the optimal parameter equals $1/\chi_f(G(\tau))$ and local Clifford measurements are sufficient. For general graphs the paper supplies SDP hierarchies and a multiplicative-weight scheme that approximate $\delta_\tau$ from above and below and construct near-optimal ensembles, with numerical evidence that the gap closes quickly on small graphs. Applied to Hamiltonian estimation, the resulting strategies give variance bounds that improve on optimized joint-measurement baselines for small molecules.

Load-bearing premise

The paper relies on a classification taken from the companion work: every Pauli anticommutation graph realizable on two qubits is $\hbar$-perfect, meaning its $\beta$ number equals its weighted independence number for all weights; this classification is cited rather than proved here, and the two-qubit Clifford-optimality claim fails if any two-qubit realization escapes it.

Editorial extensions

If this is right

  • Under any partition into two-qubit blocks, the optimal sample-complexity parameter is exactly the reciprocal fractional chromatic number of the union frustration graph, and local Clifford measurements achieve it.
  • When only single-qubit measurements are allowed, Clifford measurements reduce to Pauli measurements, so Pauli measurements are optimal in that scenario.
  • For any partition whose local frustration graphs are all $\hbar$-perfect, or whose union graph is perfect, the same exactness holds: Clifford measurements are optimal and no better strategy exists.
  • For general frustration graphs, the SDP hierarchy and multiplicative-weight method produce near-optimal measurement ensembles; on all graphs with at most seven vertices the first hierarchy level already gives the exact sample-complexity parameter.
  • For molecular Hamiltonian energy estimation, the strategy built from a single linear program yields variance bounds up to 20–200% better than the optimized joint-measurement approach used as baseline.

Reading between the lines

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

  • If the $\hbar$-perfect classification were extended from one- and two-qubit Pauli realizations to three-qubit blocks, the same proof would immediately make Clifford measurements optimal for those partitions as well.
  • The sandwich bounds suggest a practical workflow for fixed-connectivity hardware: compile the Hamiltonian under the device's partition, compute $\chi_f(G(\tau))$ and $\vartheta(\overline{G(\tau)})$ once, and use the gap as a certificate of how much is lost by the hardware constraint.
  • One open question implied by the paper is whether the lower bound $\max_{\tau}1/\chi_f(G(\tau))$ for unrestricted partition choice is ever tight; that could be probed numerically with the hierarchy on small systems.
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 studies memoryless shadow tomography of Pauli observables under block-local measurement constraints, where each copy is measured once and entangling measurements are limited to blocks of at most m qubits. The authors introduce a partition-dependent sample-complexity parameter δτ, prove a min-max reformulation over product states (Lemma 2), and show that δτ lies between the reciprocal fractional chromatic number and the reciprocal Lovász number of a certain union of local frustration graphs (Theorem 3). They further prove that Clifford measurements attain the lower bound whenever the local frustration graphs are ℏ-perfect or the union graph is perfect, and they specialize this to two-qubit blocks (Corollary 4). The paper also presents SDP hierarchies, a see-saw/multiplicative-weights algorithm for general graphs, and applies the framework to molecular Hamiltonian estimation with reported variance improvements over existing joint-measurement strategies.

Significance. If the results hold, this is a substantial contribution to shadow tomography with realistic measurement constraints. The graph-theoretic reduction is clean, the derivations are parameter-free, and the paper provides concrete algorithmic constructions together with molecular benchmarks that are derived rather than fitted. The inclusion of reproducible code and the convergence analysis of the numerical hierarchies are notable strengths. The main caveat is that one headline claim — two-qubit Clifford optimality — depends on a classification imported from a companion preprint, and one step of the main theorem's proof is stated without the supporting argument for the unconditional lower bound. These issues are local and fixable, but they need to be addressed before the claims can be taken as fully established.

major comments (2)
  1. [Corollary 4] Corollary 4 asserts that for any partition into pairs of qubits, δτ = 1/χf(G(τ)) with local Clifford optimality, citing [25] for the statement that every Pauli frustration graph on two qubits is ℏ-perfect. This classification is not proved in the present manuscript or its Supplemental Material, and [25] is a companion preprint by the same group rather than an included derivation. Because Corollary 4 is the basis for the abstract's claim of optimality in 'all two-qubit measurement scenarios' and for the two-qubit columns in Tables I and II, this dependency is load-bearing. The authors should either supply a proof of the two-qubit classification (the case is finite and likely amenable to a rank-over-F2 case analysis) or explicitly state Corollary 4 and the associated numerical results as conditional on the companion work.
  2. [SM Sec. C, proof of Theorem 3] Theorem 3 states the interval δτ ∈ [1/χf(G(τ)), 1/ϑ(G(τ))] unconditionally, but the proof in the Supplemental Material only establishes the lower bound under the additional assumption that all Gj are ℏ-perfect (leading to δτ = 1/χf(G(τ))) or that G(τ) is perfect. The main text asserts that for ℏ-imperfect Gj, '1/χf(G′) might only provide a lower bound', but no proof of this unconditional lower bound is given. Please add the missing argument (for example, using common eigenstates for each independent set of G(τ) to lower-bound the min-max expression) or restate the theorem with the hypotheses under which each bound is proved.
minor comments (4)
  1. [Main text, 'Optimal measurement strategies' / SM Sec. A B] The quantum de Finetti convergence bound is stated with different constants in the main text (4(2d_i − max_i d_i)/(m+2)) and in the Supplemental Material (8d_i/(m+2), Eqs. (24)–(25)). These should be reconciled, and the derivation should be checked against the cited theorem.
  2. [Main text, bipartition discussion] The discussion of the bipartition case first defines the frustration graph of {Si} as the XOR union of G1 and G2, then switches to the edge-union G′ for the fractional-chromatic-number bound. This transition is confusing and should be clarified explicitly, since G′ rather than the XOR graph is the graph appearing in Theorem 3.
  3. [Conclusion] There is a typo in the final paragraph: 'acounts for' should be 'accounts for'.
  4. [Reference [25]] Reference [25] is a preprint; since Corollary 4 relies on it, the authors should state its status and provide a version identifier, or avoid the dependence by proving the needed two-qubit classification in this work.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular reduction: the derivation is a chain of definitions and graph-parameter bounds; the two-qubit Clifford-optimality claim rests on an unproved companion classification, which is a verification gap rather than a circular step.

full rationale

The central chain is non-circular. Equation (2) defines delta as min_w beta(G,w), and Theorem 1 derives the interval [1/chi_f(G), 1/vartheta(bar G)] from the independent bounds alpha <= beta <= vartheta; the hbar-perfect case follows from the definition beta = alpha, with Clifford optimality imported from [14,15]. Lemma 2 and Theorem 3 derive delta_tau by minimax and BETA/STAB manipulations, and the hbar-perfect tightness argument uses the independent-set intersection structure of the G_j, not the conclusion. No parameter is fitted to a subset of data and renamed a prediction; the molecular tables compare analytic variance bounds. The only load-bearing imported fact is the companion classification 'the graph of any Pauli observables on two qubits is always hbar-perfect [25]' used in Corollary 4. Because [25] is a different, parameter-free theorem by overlapping authors and the present text gives no proof, this is a self-citation and verification concern, not an equivalence-by-construction. Accordingly, no circular step is exhibited, and the score is low.

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

The central claims rest on prior theorems from the same research group (beta-number bounds and hbar-perfectness) and on standard convex relaxations. No free parameters are fitted to data, and no new physical entities are introduced.

assumptions (5)
  • domain assumption The weighted beta number beta(G,w) is a graph parameter with alpha(G,w) <= beta(G,w) <= vartheta(G,w), and BETA(G) = STAB(G) for hbar-perfect graphs.
    Imported from the authors' prior work [24] and used in Theorem 1 and Theorem 3 without reproof.
  • domain assumption Every Pauli frustration graph realizable on one or two qubits is hbar-perfect.
    Companion work [25] supplies this classification; it is load-bearing for Corollary 4 and the single-qubit case.
  • domain assumption Clifford measurements achieve sample complexity 1/chi_f(G) for Pauli shadow tomography.
    Prior results [14,15] establish this; the paper uses it to conclude Clifford optimality when the graph parameter is tight.
  • standard math Quantum de Finetti bounds for symmetric extensions approximate separable states with trace error of order d/(m+2).
    Used in SM Section A.B to argue convergence of the SDP hierarchy; the main text and SM state different constant factors.
  • standard math Strong duality holds for the SDP relaxations via a Slater-type argument.
    Used in SM Section A.A to justify the single-SDP reformulation of omega_r.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Optimal strategies for shadow tomography with limited resources." pith.science (2026). https://pith.science/paper/3VOU74CU

@misc{pith2026260808429,
  author       = {Pith},
  title        = {Pith review of: Optimal strategies for shadow tomography with limited resources},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3VOU74CU}},
  note         = {Machine review of arXiv:2608.08429}
}
read the original abstract

Shadow tomography addresses the task of efficiently predicting many expectation values of an unknown quantum state from randomized measurements on comparatively few copies. Existing analyses promise large scaling advantages, but the optimal strategies realizing these guarantees are not always known, and the required measurements are potentially challenging to implement on current hardware. We address this gap for Pauli observables by computing optimal sample-complexity parameters and constructing optimal measurement strategies under realistic resource constraints. We focus on memoryless protocols, where each copy is measured only once, and on measurements with bounded interaction range. Our approach reduces the problem to the analysis of graph parameters of the frustration graph encoding the Pauli anticommutation relations. We provide efficient numerical methods for the general case and analytically prove that Clifford measurements are optimal in many situations. This includes all perfect graphs, all single-qubit, all two-qubit measurement scenarios, and more. Applied to Hamiltonian energy estimation, our framework yields constructive strategies and improved variance bounds for molecular benchmarks.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

79 extracted references · 65 canonical work pages

  1. [25]

    Busch, P

    P. Busch, P. Lahti, J.-P. Pellonpää, and K. Ylinen, Quantum measurement, Vol. 23 (Springer, 2016)

    work page 2016

  2. [27]

    H. Chen, Y. Chen, and H. Yuan, Incompatibility measures in multiparameter quantum estimation under hierarchical quantum measurements, Phys. Rev. A 105, 062442 (2022)

    work page 2022

  3. [1]

    For a given Hamiltonian H = Pn i=1ciSi and parti- tionτ , generate ˜ci =c2/3 i / P jc2/3 j and graph G(τ )

    work page

  4. [2]

    P I∋itI ≥T ˜ci, ∀i, P I∈ItI = 1 , tI ≥ 0, where I is the set of all maximal independent sets of G(τ )

    Solve the linear program: maxT s.t. P I∋itI ≥T ˜ci, ∀i, P I∈ItI = 1 , tI ≥ 0, where I is the set of all maximal independent sets of G(τ )

    work page

  5. [3]

    Digital Spain 2026 Agenda

    With the optimal solution {tI }, implement a sharp joint measurement of Pauli observables {Si,j} for each part j with probability tI . With this approach, N measurement rounds yield Var( ˆH) ≤λmax X i,j cicjTi,j TiTj SiSj /N ≤χf (G, ˜c) X i c2/3 i 3 /N, (9) where Ti = P I∋itI and Ti,j = P I∋i,jtI . Table I com- pares our approach with the optimized joint ...

    work page 2026

  6. [4]

    (64) The steps are as following:

    Oracle: See-saw optimisation for a fixed weight vector We firstly introduce the see-saw method to solve the inner maximization for a fixed probability vector w as an oracle, which is just the weighted beta number of the corresponding frustration graph [ 24], β(G,w ) = max ρ,σ X i wi Tr(Siρ) Tr(Siσ). (64) The steps are as following:

    work page

  7. [5]

    Initialisation: choose a random pure state ρ

    work page

  8. [6]

    Update σ: let M1 = P iwi Tr(Siρ)Si, and set σ to the pure state |ψ⟩⟨ψ| where |ψ⟩ is the eigenvector of M1 with largest eigenvalue

    work page

Show all 79 references
  1. [7]

    Update ρ: let M2 = P iwi Tr(Siσ)Si, and set ρ to the maximal eigenstate of M2

  2. [8]

    Because each step maximises a linear functional over pure states, the objective is nondecreasing and converges to a local optimum

    Repeat steps 2–3 until the objective converges, e.g., the difference between two iterations is less than 10−5. Because each step maximises a linear functional over pure states, the objective is nondecreasing and converges to a local optimum. In practice, we can run this algorit...

  3. [9]

    For a given T , set the learning rate η = p (8 lnm)/T

    Algorithm: Multiplicative weight update To solve the outer minimisation over w, we employ the multiplicative weights update (MWU) method [ 61, 62]. For a given T , set the learning rate η = p (8 lnm)/T . The algorithm proceeds as follows: • Start with uniform weights w(0) i = ...

  4. [10]

    Using the oracle to compute optimal states (ρt,σ t) for the current weight vector w(t−1)

  5. [11]

    Define ci = Tr(Siρt) Tr(Siσt)

  6. [12]

    After T iterations we collect the ensemble E = {(ρt,σ t)}T t=1

    Update w(t) i =w(t−1) i exp(−ηc i) and renormalize it as a probability distribution. After T iterations we collect the ensemble E = {(ρt,σ t)}T t=1. Standard regret analysis for MWU yields the following bound (see, e.g. [ 63] for a similar statistical context). Let w = 1 T PT ...

  7. [13]

    All such ℏ-imperfect graphs with no more than 9 vertices are already determined in a companion work [ 24]

    Results on graphs with no more than 9 vertices Since we have proven that the Clifford measurement strategy is already optimal in the case that the frustration graph is ℏ-perfect, we only need to focus on the ℏ-imperfect ones. All such ℏ-imperfect graphs with no more than 9 vert...

  8. [14]

    ForH = Pn i=1ciSi and a given partition τ , generate the graph G and compute ˜ci =c2/3 i / P jc2/3 j

  9. [15]

    ( 69) below to obtain the optimal {tI }

    Solve the linear program in Eq. ( 69) below to obtain the optimal {tI }

  10. [16]

    S” , “S1

    With these {tI }, implement a sharp joint measurement of the commuting Pauli strings in each independent set I with probability tI . The detailed derivation of the variance bounds and the linear program are as follows. Let I denote the set of all maximal independent sets of G....

  11. [17]

    Gühne, E

    O. Gühne, E. Haapasalo, T. Kraft, J.-P. Pellonpää, and R. Uola, Colloquium: Incompatible measurements in quantum information science, Rev. Mod. Phys. 95, 011003 (2023)

    2023

  12. [18]

    S. A. Yadavalli, N. Andrejic, and R. Kunjwal, Qualitative equivalence between incompatibility and Bell nonlocality, Phys. Rev. A 110, L060201 (2024)

    2024

  13. [19]

    Howard, J

    M. Howard, J. Wallman, V. Veitch, and J. Emerson, Contextuality supplies the ‘magic’ for quantum computation, Nature 510, 351 (2014)

    2014

  14. [20]

    Mannalatha, S

    V. Mannalatha, S. Mishra, and A. Pathak, A comprehensive review of quantum random number generators: concepts, classification and the origin of randomness, Quantum Inf. Process. 22, 439 (2023)

    2023

  15. [21]

    Zhang, H

    H. Zhang, H. Zhu, R. He, Y. Zhang, C. Ding, L. Hanzo, and W. Gao, Towards global quantum key distribution, Nat. Rev. Electr. Eng. 2, 806 (2025)

    2025

  16. [22]

    Y. Xiao, C. Guo, F. Meng, N. Jing, and M.-H. Yung, Incompatibility of observables as state-independent bound of uncertainty relations, Phys. Rev. A 100, 032118 (2019)

    2019

  17. [23]

    Schwonnek, Uncertainty Relations in Quantum Theory , Ph.D

    R. Schwonnek, Uncertainty Relations in Quantum Theory , Ph.D. thesis , Leibniz Universität Hannover (2018)

    2018

  18. [24]

    L. E. Ballentine, The uncertainty principle and the statistical interpretation of quantum mechanics, Can. J. Phys. 47, 2417 (1969)

    1969

  19. [26]

    Y.-L. Mao, H. Chen, C. Niu, Z.-D. Li, S. Yu, and J. Fan, Testing Heisenberg-type measurement uncertainty relations of three observables, Phys. Rev. Lett. 131, 150203 (2023)

    2023

  20. [28]

    Anshu, S

    A. Anshu, S. Arunachalam, T. Kuwahara, and M. Soleimanifar, Sample-efficient learning of interacting quantum systems, Nat. Phys. 17, 931 (2021)

    2021

  21. [29]

    Huang, R

    H.-Y. Huang, R. Kueng, and J. Preskill, Predicting many properties of a quantum system from very few measurements, Nat. Phys. 16, 1050 (2020)

    2020

  22. [30]

    S. Chen, W. Gong, and Q. Ye, Optimal tradeoffs for estimating Pauli observables, in 2024 IEEE 65th Annual Symposium on Foundations of Computer Science (FOCS) (IEEE, 2024) pp. 27–30

    2024

  23. [31]

    R. King, D. Gosset, R. Kothari, and R. Babbush, Triply efficient shadow tomography, PRX Quantum 6, 010336 (2025)

    2025

  24. [32]

    S. Chen, J. Cotler, H.-Y. Huang, and J. Li, Exponential separations between learning with and without quantum memory, 18 in 2021 IEEE 62nd Annual Symposium on Foundations of Computer Science (FOCS) (IEEE, 2022) pp. 574–585

    2021

  25. [33]

    Aaronson, Shadow tomography of quantum states, SIAM J

    S. Aaronson, Shadow tomography of quantum states, SIAM J. Comput. (2020)

    2020

  26. [34]

    H. C. Nguyen, J. L. Bönsel, J. Steinberg, and O. Gühne, Optimizing shadow tomography with generalized measurements, Phys. Rev. Lett. 129, 220502 (2022)

    2022

  27. [35]

    Rozon, N

    P.-G. Rozon, N. Bao, and K. Agarwal, Optimal twirling depth for classical shadows in the presence of noise, Phys. Rev. Lett. 133, 130803 (2024)

    2024

  28. [36]

    Gottesman, Stabilizer Codes and Quantum Error Correction , Ph.D

    D. Gottesman, Stabilizer Codes and Quantum Error Correction , Ph.D. thesis, California Institute of Technology (1997), arXiv:quant-ph/9705052

    1997 arXiv

  29. [37]

    Loizeau, J

    N. Loizeau, J. C. Peacock, and D. Sels, Quantum many-body simulations with PauliStrings.jl, SciPost Phys. Codebases , 054 (2025)

    2025

  30. [38]

    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

  31. [39]

    Preskill, Quantum computing in the NISQ era and beyond, Quantum 2, 79 (2018) , 1801.00862v3

    J. Preskill, Quantum computing in the NISQ era and beyond, Quantum 2, 79 (2018) , 1801.00862v3

    2018 arXiv

  32. [40]

    Z.-P. Xu, R. Schwonnek, and A. Winter, Bounding the joint numerical range of pauli strings by graph parameters, PRX Quantum 5, 020318 (2024)

    2024

  33. [41]

    Z.-P. Xu, J. Wang, Q. Ye, G. KoSSmann, R. Schwonnek, and A. Winter, Simultaneous variances of pauli strings, weighted independence numbers, and a new kind of perfection of graphs (2025), arXiv:2511.13531 [quant-ph]

    2025

  34. [42]

    Bharti, A

    K. Bharti, A. Cervera-Lierta, T. H. Kyaw, T. Haug, S. Alperin-Lea, A. Anand, M. Degroote, H. Heimonen, J. S. Kottmann, T. Menke, W.-K. Mok, S. Sim, L.-C. Kwek, and A. Aspuru-Guzik, Noisy intermediate-scale quantum algorithms, Rev. Mod. Phys. 94, 015004 (2022)

    2022

  35. [43]

    McNulty, F

    D. McNulty, F. B. Maciejewski, and M. Oszmaniec, Estimating quantum hamiltonians via joint measurements of noisy noncommuting observables, Phys. Rev. Lett. 130, 100801 (2023)

    2023

  36. [44]

    M. A. Nielsen and I. L. Chuang, Quantum Computation and Quantum Information: 10th Anniversary Edition (Cambridge University Press, Cambridge, England, UK, 2010)

    2010

  37. [45]

    Chapman and S

    A. Chapman and S. T. Flammia, Characterization of solvable spin models via graph invariants, Quantum 4, 278 (2020) , 2003.05465v2

    2020 arXiv

  38. [46]

    de Gois, K

    C. de Gois, K. Hansenne, and O. Gühne, Uncertainty relations from graph theory, Phys. Rev. A 107, 062211 (2023)

    2023

  39. [47]

    M. B. Hastings and R. O’Donnell, Optimizing strongly interacting fermionic Hamiltonians, in ACM Conferences (Associ- ation for Computing Machinery, New York, NY, USA, 2022) pp. 776–789

    2022

  40. [48]

    Y. S. Samoilenko, Spectral Theory of Families of Self-Adjoint Operators (Springer Netherlands, Dordrecht, The Netherlands, 1991)

    1991

  41. [49]

    M. B. Morán and F. Huber, Uncertainty relations from state polynomial optimization, Phys. Rev. Lett. 132, 200202 (2024)

    2024

  42. [50]

    Bäumer, N

    E. Bäumer, N. Gisin, and A. Tavakoli, Demonstrating the power of quantum computers, certification of highly entangled measurements and scalable quantum nonlocality, npj Quantum Inf. 7, 117 (2021)

    2021

  43. [51]

    Bianchi, C

    L. Bianchi, C. Marconi, and D. Bacco, Bell state measurements in quantum optics: a review of recent progress and open challenges, Quantum Sci. Technol. 11, 023001 (2026)

    2026

  44. [52]

    Welte, P

    S. Welte, P. Thomas, L. Hartung, S. Daiss, S. Langenfeld, O. Morin, G. Rempe, and E. Distante, A nondestructive Bell-state measurement on two distant atomic qubits, Nat. Photonics 15, 504 (2021)

    2021

  45. [53]

    Zhang, J

    H. Zhang, J. Ruan, D. Tullsen, Y. Ding, A. Li, and T. Humble, Oneadapt: Resource-adaptive compilation of measurement- based quantum computing for photonic hardware, in MICRO 2025 - 58th IEEE/ACM International Symposium on Mi- croarchitecture, Proceedings of the Annual Internat...

    2025

  46. [54]

    Aghaee Rad et al

    H. Aghaee Rad et al. , Scaling and networking a modular photonic quantum computer, Nature 638, 912 (2025)

    2025

  47. [55]

    Raussendorf and H

    R. Raussendorf and H. J. Briegel, A one-way quantum computer, Phys. Rev. Lett. 86, 5188 (2001)

    2001

  48. [56]

    Bravyi and A

    S. Bravyi and A. Kitaev, Universal quantum computation with ideal Clifford gates and noisy ancillas, Phys. Rev. A 71, 022316 (2005)

    2005

  49. [57]

    Kielpinski, C

    D. Kielpinski, C. Monroe, and D. J. Wineland, Architecture for a large-scale ion-trap quantum computer, Nature 417, 709 (2002)

    2002

  50. [58]

    J. M. Pino, J. M. Dreiling, C. Figgatt, J. P. Gaebler, S. A. Moses, M. S. Allman, C. H. Baldwin, M. Foss-Feig, D. Hayes, K. Mayer, C. Ryan-Anderson, and B. Neyenhuis, Demonstration of the trapped-ion quantum CCD computer architecture, Nature 592, 209 (2021)

    2021

  51. [59]

    J. You, J. Tsai, and F. Nori, Quantum computing with many superconducting qubits, in New Directions in Mesoscopic Physics (Towards Nanoscience) (Springer, 2003) pp. 351–360

    2003

  52. [60]

    Horodecki, P

    M. Horodecki, P. Horodecki, and R. Horodecki, Separability of mixed states: necessary and sufficient conditions, Phys. Lett. A 223, 1 (1996)

    1996

  53. [61]

    Peres, Separability criterion for density matrices, Phys

    A. Peres, Separability criterion for density matrices, Phys. Rev. Lett. 77, 1413 (1996)

    1996

  54. [62]

    Størmer, Symmetric states of infinite tensor products of C ∗-algebras, J

    E. Størmer, Symmetric states of infinite tensor products of C ∗-algebras, J. Funct. Anal. 3, 48 (1969)

    1969

  55. [63]

    C. M. Caves, C. A. Fuchs, and R. Schack, Unknown quantum states: The quantum de Finetti representation, J. Math. Phys. 43, 4537 (2002)

    2002

  56. [64]

    Christandl, R

    M. Christandl, R. König, G. Mitchison, and R. Renner, One-and-a-half quantum de Finetti theorems, Commun. Math. Phys. 273, 473 (2007)

    2007

  57. [65]

    I. Klep, V. Magron, J. Volčič, and J. Wang, State polynomials: positivity, optimization and nonlinear Bell inequalities, Mathematical Programming 207, 645 (2024)

    2024

  58. [66]

    Hadfield, S

    C. Hadfield, S. Bravyi, R. Raymond, and A. Mezzacapo, Measurements of quantum hamiltonians with locally-biased 19 classical shadows, Commun. Math. Phys. 391, 951 (2022)

    2022

  59. [67]

    Hadfield, Adaptive Pauli shadows for energy estimation (2021), arXiv:2105.12207 [quant-ph]

    C. Hadfield, Adaptive Pauli shadows for energy estimation (2021), arXiv:2105.12207 [quant-ph]

    2021 arXiv

  60. [68]

    Korhonen, S

    K. Korhonen, S. Mangini, J. Malmi, H. Vappula, and D. Cavalcanti, Improving shadow estimation with locally-optimal dual frames (2026), arXiv:2511.02555 [quant-ph]

    2026 arXiv

  61. [69]

    Peruzzo, J

    A. Peruzzo, J. McClean, P. Shadbolt, M.-H. Yung, X.-Q. Zhou, P. J. Love, A. Aspuru-Guzik, and J. L. O’Brien, A variational eigenvalue solver on a photonic quantum processor, Nat. Commun. 5, 4213 (2014)

    2014

  62. [70]

    Kandala, A

    A. Kandala, A. Mezzacapo, K. Temme, M. Takita, M. Brink, J. M. Chow, and J. M. Gambetta, Hardware-efficient variational quantum eigensolver for small molecules and quantum magnets, Nature 549, 242 (2017)

    2017

  63. [71]

    P. J. J. O’Malley et al. , Scalable quantum simulation of molecular energies, Phys. Rev. X 6, 031007 (2016)

    2016

  64. [72]

    Hempel, C

    C. Hempel, C. Maier, J. Romero, J. McClean, T. Monz, H. Shen, P. Jurcevic, B. P. Lanyon, P. Love, R. Babbush, A. Aspuru-Guzik, R. Blatt, and C. F. Roos, Quantum chemistry calculations on a trapped-ion quantum simulator, Phys. Rev. X 8, 031022 (2018)

    2018

  65. [73]

    Josz and D

    C. Josz and D. Henrion, Strong duality in Lasserre’s hierarchy for polynomial optimization, Optim. Lett. 10, 3 (2016)

    2016

  66. [74]

    F. G. S. L. Brandão and M. Christandl, Detection of Multiparticle Entanglement: Quantifying the Search for Symmetric Extensions, Phys. Rev. Lett. 109, 160502 (2012)

    2012

  67. [75]

    Schrijver, Fractional packing and covering, in Packing and Covering in Combinatorics , Vol

    A. Schrijver, Fractional packing and covering, in Packing and Covering in Combinatorics , Vol. 106 (Mathematisch Centrum Amsterdam, 1979) pp. 201–274

    1979

  68. [76]

    D. E. Knuth, The Sandwich Theorem, Electron. J. Comb. 1, A1 (1994)

    1994

  69. [77]

    Arora, E

    S. Arora, E. Hazan, and S. Kale, The multiplicative weights update method: a meta-algorithm and applications, Theory of Computing 8, 121 (2012)

    2012

  70. [78]

    Freund and R

    Y. Freund and R. E. Schapire, Adaptive game playing using multiplicative weights, Games and Economic Behavior 29, 79 (1999)

    1999

  71. [79]

    A. A. Fernández, J. Blanchet, J. L. M. Olea, C. Qiu, J. Stoye, and L. Tan, Epsilon-Minimax Solutions of Statistical Decision Problems, arXiv 10.48550/arXiv.2509.08107 (2025), 2509.08107

    2025 doi

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.