Pith. sign in

REVIEW 2 major objections 5 minor 62 references

This paper proves that fixed stabilizer circuits cannot be helped by any ancilla, while adaptive stabilizer circuits with a single non-stabilizer qubit ancilla can improve quantum state discrimination, and derives the exact qubit success pr

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-01 09:09 UTC pith:P65AZPHV

load-bearing objection Solid paper, conditional central formula: Result 3's exact adaptive success probability rests on an unverified sign-pattern achievability step; otherwise the results and SDP framework are worth taking seriously. the 2 major comments →

arxiv 2607.27339 v1 pith:P65AZPHV submitted 2026-07-29 quant-ph

Quantum State Discrimination With Stabilizer Circuits

Benjamin Stratton , Santiago Zamora , Rafael Chaves This is my paper
classification quant-ph
keywords quantum state discriminationstabilizer circuitsnon-stabilizernessmagic statesadaptive measurementsHelstrom boundsemidefinite programmingquantum random access codes
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

Quantum state discrimination asks how well two states can be told apart when only certain measurements are allowed. This paper restricts measurements to stabilizer circuits—Clifford unitaries plus computational-basis measurements—and asks how much discrimination power is lost, and how much a bit of 'magic' (non-stabilizerness) can restore. The central result is an operational separation: fixed stabilizer circuits gain nothing from any ancilla, whether stabilizer or not, but adaptive stabilizer circuits do. For single-qubit systems the paper gives an exact formula for the adaptive success probability in terms of the Bloch-vector difference of the two states and the Bloch vector of the ancilla, showing that an H-type magic state is always the best ancilla. Beyond that, it recasts discrimination with non-stabilizer measurements as a semidefinite program, derives a closed-form bound interpolating between the stabilizer limit and the Helstrom limit, and applies the results to quantum random access codes and unitary synthesis with a finite number of magic states.

Core claim

The paper's core discovery is a sharp dichotomy. For fixed stabilizer circuits, appending arbitrary ancillas—even non-stabilizer ones—never improves the minimum-error success probability; the stabilizer-only success probability P = 1/2 + (1/4)||Δ||∞ is universal. With adaptivity, a single qubit ancilla with Bloch vector n_ω can help, and for qubits the exact optimal success probability is P = 1/2 + (1/4)max{||Δ||∞, ||Δ||2−ky ||n_ω||2−ky /2}, where the 2−ky norm sums the two largest absolute components. This formula implies that the best ancilla is any H-type magic state, and that the Helstrom bound is recovered only for pairs of orthogonal stabilizer states or orthogonal H-type magic states.

What carries the argument

The central object is the difference operator Δ=ρ−σ, which fully determines the discrimination problem. The paper's arguments hinge on rewriting any measurement achievable by a stabilizer circuit as a coarse-grained stabilizer basis, then using a lemma (Lemma 2) that reduces the optimal success probability to half of the sum of the absolute traces of Δ against the POVM elements. For the adaptive qubit case, the key identity is that the advantage factorizes into a product of two second vector KY-Fan norms—the sum of the two largest absolute components of the Bloch difference and of the ancilla—and the proof bounds this product by optimizing over entangled stabilizer bases. For non-stabilizer

Load-bearing premise

Two steps carry the quantitative results: Result 3 assumes without proof that every sign configuration of the Bloch components Δ_a, Δ_b, n_α, n_β is realizable by some Clifford choice (SM H5), and Result 4 assumes the no-leakage ansatz, which the paper itself shows can fail (Fig. 3); if either assumption breaks, the corresponding formula is an upper bound, not the exact or tight value.

What would settle it

For Result 3: enumerate all commuting Pauli pairs on two qubits and all single-qubit Clifford choices V0,V1, and check for random Δ and ω whether the closed-form value equals the maximum over circuits; if any state pair yields a strict gap, the exhaustive sign-pattern claim is false. For Result 4: run the SDP with and without the no-leakage constraint on a grid of asymmetric Δ and compare—the paper's Fig. 3 already shows a gap, so a closed-form measure of the leakage term would settle how far the bound is from exact.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • Fixed stabilizer circuits define an intrinsic discrimination limit: no amount of ancilla preparation can push past ||Δ||∞, so any advantage must come from adaptivity or from non-stabilizer measurements.
  • Adaptive circuits with one H-type magic state strictly outperform fixed circuits for state pairs whose Bloch difference has two comparable large components, and the advantage is exactly quantified by the product of the two KY-Fan norms.
  • The exact qubit formula doubles as a benchmark for magic-state cost: a target success probability can be converted into the required non-stabilizerness of the ancilla.
  • The SDP formulation, though not efficient for many qubits, gives a universal method to compute optimal discrimination under bounded non-stabilizerness in any dimension.
  • The random-access-code result shows that optimal n→1 QRACs need non-stabilizerness on the measurement side, not just in preparation.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same coarse-graining technique could be applied to other restricted measurement classes, such as Gaussian measurements in continuous variables, to see whether the fixed-vs-adaptive separation is universal.
  • Because the paper shows Result 4 is not tight when the measurement vector leaks out of the active subspace, a promising follow-up is to compute the leakage term as a function of asymmetry; that would turn the lower bound into an exact formula for asymmetric state pairs.
  • The exact qubit formula suggests a concrete protocol: use one H-type magic state as a resource to rotate the measurement basis, and the success probability is simply the product of the two 'top-two' norms; this could be tested directly in a quantum processor with a single magic-state injection.
  • The sign-pattern enumeration in SM H5, if automated rather than asserted, would certify Result 3 numerically for all states; such a certification would be a natural benchmark for the proof's completeness.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 5 minor

Summary. The paper studies minimum-error quantum state discrimination when the measurement is restricted to stabilizer circuits, possibly augmented by ancillas. It reports four main results: (i) fixed stabilizer circuits with arbitrary stabilizer ancillas do not improve the success probability (Result 1); (ii) for qubits, fixed stabilizer circuits with one arbitrary qubit ancilla also give no improvement (Result 2); (iii) for qubits, adaptive stabilizer circuits with one arbitrary qubit ancilla achieve the exact success probability 1/2 + (1/4)max{||Δ||∞,||Δ||_{2-ky}||n_ω||_{2-ky}/2} (Result 3); and (iv) when non-stabilizerness is introduced directly into the POVM, the problem can be cast as an SDP and analytical lower bounds are derived (Result 4). Applications to quantum random access codes and unitary synthesis fidelity are also presented (Results 5 and 6). The proofs of Results 1–3 are in the Supplementary Material, while Result 4 rests on a no-leakage ansatz whose limitations are acknowledged in Fig. 3.

Significance. If the results are correct, the paper provides an operational characterization of non-stabilizerness (magic) in a fundamental information-theoretic task, showing that adaptive stabilizer circuits with a non-stabilizer ancilla strictly outperform fixed stabilizer circuits and quantifying the gap via the 2nd Ky Fan norm. The SDP formulation and the exact qubit formula are concrete tools that could be used in further resource-theoretic analyses. Strengths include explicit proofs for Results 1–3 in the SM, a reproducible SDP code, and falsifiable numerical predictions in Fig. 3. However, the exact formula of Result 3 relies on an unproduced finite case analysis, and Result 4's statement overclaims tightness; both issues need to be resolved before the paper can be accepted.

major comments (2)
  1. [SM H5, after Eq. (H110)] The proof of Result 3 derives an upper bound via the triangle inequality and then asserts achievability by 'an exhaustive consideration of all possible sign combinations' without supplying the enumeration. This step is load-bearing: if some sign configuration of Δ_a, Δ_b, n_α, n_β were not realizable by a valid choice of Clifford unitaries U, V0, V1, the closed form in Result 3 would overestimate the adaptive success probability. The same paragraph also uses the identity sign(P_iP_j) = -sign(P_kP_j) without derivation. The claim is plausible and can be proven by a short sign argument (choosing the common sign to make the two terms add constructively), but as written the proof is incomplete. Please provide the missing case analysis or a direct argument, and prove the sign identity.
  2. [Result 4 and Fig. 3] Result 4 states that the success probability is 'tightly bounded below' by max_A P^A_suc(µ). However, the paper's own Fig. 3 and the following paragraph show that for asymmetric trajectories (e.g., |T⟩ → |H⟩) the SDP solution strictly exceeds the analytical bound at intermediate µ, so the bound is not universally tight. The theorem as stated overclaims; it should say 'bounded below', with tightness stated only under the symmetry conditions identified in SM H6 (e.g., Δ aligned with a symmetry axis of the stabilizer octahedron). This is a statement-level inconsistency that should be corrected.
minor comments (5)
  1. [SM F1, Eq. (F8)] The explanatory chain leading to the SDP formulation is confusing: the 'guess' terminology suggests suboptimality, but the variables σ'_i and X_i are actual optimization variables in the final SDP. Please rewrite this passage to distinguish the definitional SDP representation of the 1-norm from the feasible variables enforcing the magic constraint.
  2. [Appendix D] The sentence 'Due to the symmetry of these states, Pµ_suc = max_A P^A_suc(µ). Therefore it also constitutes the solution obtained through the SDP' is ambiguous. It should clarify that for the symmetric geodesic the analytical bound equals the SDP optimum, not that the SDP is being replaced.
  3. [Main text, abstract] The abstract says 'adaptive circuits do' provide additional discrimination power with stabilizer ancillas. This is established only by the explicit two-qubit example in Appendix C, not as a general theorem; the phrasing should make this clear.
  4. [Result 3, notation] The phrase 'the 2nd vector KY-Fan norm' is nonstandard; it should be defined explicitly as the sum of the two largest absolute components, which it is, but consider using a more conventional notation to avoid confusion with the 2-norm.
  5. [Code availability] The GitLab link [33] is cited but not included in the reference list with a URL. Since the SDP results and the no-leakage restricted SDP are important reproducibility artifacts, please provide the full link.

Circularity Check

0 steps flagged

No significant circularity: the derivations are self-contained and do not reduce to fitted inputs or load-bearing self-citations.

full rationale

I walked the derivation chain from the POVM formalism through the stabilizer-circuit characterizations (Appendix E), the fixed-circuit results (SM H1, H3), the adaptive qubit formula (SM H5), the non-stabilizer-measurement SDP and bound (SM F1, H6), and the applications (Results 5 and 6). The central quantities are computed directly from the Pauli decomposition of Δ and the Bloch vector of the ancilla; no target success probability is assumed as an input. Lemma 1 follows from optimizing over single-qubit stabilizer measurements; Result 2 uses Lemma 5 for two-qubit stabilizer bases and the rank-one matrix (Δ)(n)^t; Result 3 starts from the adaptive instrument and derives both the triangle-inequality upper bound and an achievability claim from Clifford conjugation freedom. The unproduced sign-pattern enumeration in SM H5 is a rigor gap, not circularity: it does not import the claimed result, it asserts that certain Clifford choices realize the bound. Result 4 is explicitly a lower bound under a stated no-leakage ansatz, and Fig. 3 shows it is sometimes strict; this is an accuracy caveat, not a reduction to assumptions. The self-citations [18,31] supply the magic monotone M(ρ)=min_{σ∈STAB} 1/2||ρ−σ||_1, but the C_N surface and the SDP constraints are re-derived from the optimization variables in SM H6 and SM F1; they are not assumed in the form needed for the result. Reference [19] is used only for comparison in QRACs, not as a premise of the derivation. No fitted parameter is renamed as a prediction, and no uniqueness theorem or prior ansatz is invoked to force the main formulas. The central claims therefore have independent mathematical content, and the paper is not circular; any concerns about the omitted sign-case proof or the wording of Result 4 belong to correctness/rigor assessment, not circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 5 axioms · 0 invented entities

No parameters are fitted and no new entities are postulated. The paper's input is the standard stabilizer formalism plus an external resource (non-stabilizer ancillas or bounded-magic POVMs). The only ad hoc elements are the no-leakage/subspace ansatz in the analytical bound, which the authors themselves show is not universally tight, and the unexhibited finite sign check in the proof of Result 3.

axioms (5)
  • standard math Standard stabilizer formalism: Clifford unitaries, stabilizer states, and stabilizer POVMs are characterized by Pauli-string statistics.
    Used throughout Sections II-III and SM E.
  • domain assumption The resource measure M(ρ)=min_{σ∈STAB} 1/2||ρ-σ||_1 is a valid measure of non-stabilizerness and can be extended to unnormalized POVM elements by normalizing.
    Adopted from Refs [18,31]; used in Result 4 and SM F.
  • ad hoc to paper No-leakage ansatz: the optimal measurement Bloch vector lies in the same subspace as the closest stabilizer state; used to derive the closed-form P^A_suc in Result 4.
    Explicitly introduced in SM H6; the paper shows it fails for asymmetric geodesics (Fig. 3), making Result 4 a lower bound rather than a tight characterization.
  • ad hoc to paper Result 3 achievability relies on an unshown exhaustive sign-configuration check for the Clifford basis choices.
    SM H5 asserts 'one can do an exhaustive consideration...' without writing it; this is a proof obligation rather than a derived lemma.
  • domain assumption Adaptivity is modelled only as a measurement on the ancillary system that selects a Clifford unitary on the primary system; direct adaptive measurements on the primary system are excluded.
    Defined in Section II and SM E4; Appendix I notes a perfectly-distinguishing strategy exists if direct adaptivity on the primary system is allowed, so the separation result is model-dependent.

pith-pipeline@v1.3.0-daily-deepseek · 35259 in / 21819 out tokens · 210441 ms · 2026-08-01T09:09:33.541682+00:00 · methodology

0 comments
read the original abstract

The task of quantum state discrimination provides an operational characterization of distinguishability and plays a central role in quantum information science. Since the achievable success probability in quantum state discrimination depends both on the states being discriminated and on the allowed measurements, it is natural to study discrimination under physically motivated measurement constraints. Here, we investigate minimum-error quantum state discrimination under measurements implementable by both fixed and adaptive stabilizer circuits. Given arbitrary stabilizer-state ancillas, we show that fixed stabilizer circuits provide no additional discrimination power, whereas adaptive circuits do. Then, focusing on single-qubit systems, we derive analytical expressions for the success probability across both circuit classes when supplied with arbitrary (non-stabilizer) ancillas, showing that the performance gap between fixed and adaptive circuits persists. We further consider discrimination with non-stabilizerness incorporated directly into the measurement operators. Here, the optimization is formulated as a semidefinite program, and analytical bounds interpolating between the stabilizer and Helstrom limits for qubits are derived. Finally, we illustrate applications in quantum random access codes and bounds on unitary synthesis fidelity with a finite number of magic states.

Figures

Figures reproduced from arXiv: 2607.27339 by Benjamin Stratton, Rafael Chaves, Santiago Zamora.

Figure 1
Figure 1. Figure 1: FIG. 1. A graphical depiction of the task of QSD, where the [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. The minimum error success probability for all [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. The regions for which access to an [PITH_FULL_IMAGE:figures/full_fig_p013_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. The regions for which access to a state [PITH_FULL_IMAGE:figures/full_fig_p013_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6. The regions for which access to a state [PITH_FULL_IMAGE:figures/full_fig_p014_6.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

62 extracted references · 2 linked inside Pith

  1. [1]

    C. W. Helstrom,Quantum Detection and Estimation Theory (Academic Press, 1976)

    1976

  2. [2]

    Hence, these are the only orthogonal pure states for which Hel- strom’s bound is recovered given access to a singleH- type magic state

    The former case corresponds to when ⃗∆is the difference of two or- thogonal pure stabilizer states; the latter occurs when |⃗∆↓|= (2/ √ 2, 2/ √ 2, 0)t, meaning that ⃗∆is the differ- ence of two orthogonalH-type magic states. Hence, these are the only orthogonal pure states for which Hel- strom’s bound is recovered given access to a singleH- type magic sta...

  3. [3]

    Bae and L.-C

    J. Bae and L.-C. Kwek, Journal of Physics A: Mathematical and Theoretical48, 083001 (2015). 7

    2015

  4. [4]

    C. H. Bennett and G. Brassard, inProceedings of the IEEE International Conference on Computers, Systems and Signal Processing(Bangalore, India, 1984) pp. 175–179

    1984

  5. [5]

    S. P . et al., Advances in Optics and Photonics12, 1012 (2020)

    2020

  6. [6]

    K. M. R. Audenaert, M. Nussbaum, A. Szkoła, and F. Ver- straete, Communications in Mathematical Physics279, 251 (2008)

    2008

  7. [7]

    Zhuang, Z

    Q. Zhuang, Z. Zhang, and J. H. Shapiro, Physical review letters118, 040801 (2017)

    2017

  8. [8]

    Zhuang and S

    Q. Zhuang and S. Pirandola, Physical Review Letters125, 080505 (2020)

    2020

  9. [9]

    Chitambar and G

    E. Chitambar and G. Gour, Rev. Mod. Phys.91, 025001 (2019)

    2019

  10. [10]

    Gour,Quantum Resource Theories(Cambridge Univer- sity Press, 2025)

    G. Gour,Quantum Resource Theories(Cambridge Univer- sity Press, 2025)

    2025

  11. [11]

    Bravyi and A

    S. Bravyi and A. Kitaev, Phys. Rev. A71, 022316 (2005)

    2005

  12. [12]

    Veitch, S

    V . Veitch, S. A. Hamed Mousavian, D. Gottesman, and J. Emerson, New Journal of Physics16, 013009 (2014)

    2014

  13. [13]

    Howard and E

    M. Howard and E. Campbell, Phys. Rev. Lett.118, 090501 (2017)

    2017

  14. [14]

    Gottesman, arXiv:quant-ph/9807006 (1998)

    D. Gottesman, arXiv:quant-ph/9807006 (1998)

    Pith/arXiv arXiv 1998

  15. [15]

    Aaronson and D

    S. Aaronson and D. Gottesman, Phys. Rev. A70, 052328 (2004)

    2004

  16. [16]

    Gottesman,Stabilizer codes and quantum error correction (California Institute of Technology, 1997)

    D. Gottesman,Stabilizer codes and quantum error correction (California Institute of Technology, 1997)

    1997

  17. [17]

    E. T. Campbell, B. M. Terhal, and C. Vuillot, Nature549, 172 (2017)

    2017

  18. [18]

    A. Gu, L. Leone, S. Ghosh, J. Eisert, S. F. Yelin, and Y. Quek, Phys. Rev. Lett.132, 210602 (2024)

    2024

  19. [19]

    C. Cao, G. Cheng, A. Hamma, L. Leone, W. Munizzi, and S. F. Oliviero, PRX Quantum6, 040375 (2025)

    2025

  20. [20]

    Zamora, R

    S. Zamora, R. A. Mac ˆedo, T. S. Sarubi, M. Alves, D. Poderini, and R. Chaves, Physical Review A112, 042410 (2025)

    2025

  21. [21]

    de Oliveira Junior, R

    A. de Oliveira Junior, R. A. Mac ˆedo, J. Czartowski, J. B. Brask, and R. Chaves, Phys. Rev. A113, 052421 (2026)

    2026

  22. [22]

    Quantum state designs via magic tele- portation,

    H. L ´oio, G. Lami, L. Leone, M. McGinley, X. Turkeshi, and J. D. Nardis, “Quantum state designs via magic tele- portation,” (2026), arXiv:2510.13950 [quant-ph]

    arXiv 2026

  23. [23]

    Beverland, E

    M. Beverland, E. Campbell, M. Howard, and V . Kliuch- nikov, Quantum Science and Technology5, 035009 (2020)

    2020

  24. [24]

    J. R. Seddon, B. Regula, H. Pashayan, Y. Ouyang, and E. T. Campbell, PRX Quantum2, 010345 (2021)

    2021

  25. [25]

    Leone, S

    L. Leone, S. F. E. Oliviero, and A. Hamma, Phys. Rev. Lett. 128, 050402 (2022)

    2022

  26. [26]

    Sonya Tarabunga, M

    P . Sonya Tarabunga, M. Frau, T. Haug, E. Tirrito, and L. Piroli, Quantum Science and Technology10, 045026 (2025)

    2025

  27. [27]

    Warmuz, E

    K. Warmuz, E. Dokudowiec, C. Radhakrishnan, and T. Byrnes, Phys. Rev. Lett.135, 010203 (2025)

    2025

  28. [28]

    Predicting magic from very few measurements,

    J. M. Varela, L. L. Keller, A. de Oliveira Junior, D. A. Mor- eira, R. Chaves, and R. A. Mac ˆedo, “Predicting magic from very few measurements,” (2026), arXiv:2602.18939 [quant-ph]

    arXiv 2026

  29. [29]

    Liu and A

    Z.-W. Liu and A. Winter, PRX Quantum3, 020333 (2022)

    2022

  30. [30]

    Nonstabilizerness without magic: Classi- cally simulatable quantum states that are indistinguish- able by classically simulatable quantum circuits,

    H. Kwon, “Nonstabilizerness without magic: Classi- cally simulatable quantum states that are indistinguish- able by classically simulatable quantum circuits,” (2025), arXiv:2509.25790 [quant-ph]

    arXiv 2025

  31. [31]

    M. A. Nielsen and I. L. Chuang,Quantum Computation and Quantum Information: 10th Anniversary Edition(Cam- bridge University Press, 2010)

    2010

  32. [32]

    Palhares, S

    A. Palhares, S. Zamora, R. A. Mac ˆedo, T. S. Sarubi, J. M. Varela, G. W. Rocha, D. A. Moreira, and R. Chaves, Physics Letters A576, 131417 (2026)

    2026

  33. [33]

    Quantum random access codes with shared random- ness,

    A. Ambainis, D. Leung, L. Mancinska, and M. Ozols, “Quantum random access codes with shared random- ness,” (2009), arXiv:0810.2937 [quant-ph]

    Pith/arXiv arXiv 2009

  34. [34]

    Code for: Quan- tum state discrimination with stabilizer circuits,

    B. Stratton, S. Zamora, and R. Chaves, “Code for: Quan- tum state discrimination with stabilizer circuits,” (2026)

    2026

  35. [35]

    Matthews, S

    W. Matthews, S. Wehner, and A. Winter, Communica- tions in Mathematical Physics291, 813–843 (2009)

    2009

  36. [36]

    K. Shoemake, inProceedings of the 12th Annual Confer- ence on Computer Graphics and Interactive Techniques, SIG- GRAPH ’85 (Association for Computing Machinery, New York, NY, USA, 1985) p. 245–254

    1985

  37. [37]

    Skrzypczyk and D

    P . Skrzypczyk and D. Cavalcanti,Semidefinite Program- ming in Quantum Information Science, 2053-2563 (IOP Pub- lishing, 2023)

    2053

  38. [38]

    Howard, J

    M. Howard, J. Wallman, V . Veitch, and J. Emerson, Na- ture510, 351 (2014). 8 Supplementary Material: State Discrimination With Stabilizer Measurements CONTENTS E. Measurements Under Stabilizer Circuits 8

    2014

  39. [39]

    Two tonQubits 9 3.nQubits With Ancilla 9

  40. [40]

    Non-stabilizer measurements 10

    Adaptive Measurements 10 F. Non-stabilizer measurements 10

  41. [41]

    When does Non-stabilizerness Provide an Advantage Under Adaptive Stabilizer Circuits 12 H

    n-qudits minimum error success probability via an SDP 10 G. When does Non-stabilizerness Provide an Advantage Under Adaptive Stabilizer Circuits 12 H. Proofs 13

  42. [42]

    Proof of Result 1 13

  43. [43]

    Proof of Result 2 17

  44. [44]

    Proof of Qubit Helstrom Bound 21

  45. [45]

    Proof of Result 3 22

  46. [46]

    Proof of Result 4 27

  47. [47]

    Proof of Result 5 31

  48. [48]

    Proof of Result 6 32

  49. [49]

    Proof of Lemma 2 33 I. QSD under Adaptive Stabilizer Circuits 34 Appendix E: Measurements Under Stabilizer Circuits Measurements are modelled by Positive Operator Value Measures (POVMs), which are a set of operators{M i}i such that Mi ≥0∀iand ∑ i Mi =I. (E1) Stabilizer circuits are the subset of the stabilizer operations, and consist of Clifford unitaries...

  50. [50]

    Single Qubit For a single qubit, the POVM{|0⟩⟨0|,|1⟩⟨1|}captures a measurement in the computational basis. This is a two outcome projective measurement which can alternatively be written as: Πa = I+ (−1)aZ 2 :a∈ {0, 1} , (E2) whereZ=|0⟩⟨0| − |1⟩⟨1|is the Pauli Z operator anda∈ {0, 1}is the single bit output from performing the computa- tional basis measur...

  51. [51]

    We will first do this explicitly for two qubits, before generalising tonqubits

    Two tonQubits The above notion can easily be expanded to multiple qubits. We will first do this explicitly for two qubits, before generalising tonqubits. Measuring multiple qubits in the computational basis means measuring the single qubit computational basis mea- surement POVM in each qubit subspace. For two qubits this corresponds to measuring the four ...

  52. [52]

    We also consider the possibility of the player performingadaptive stabilizer circuits

    Adaptive Measurements The above sets of measurements achievable with stabilizer circuits are fixed measurements i.e., they are deter- mined before the circuit begins. We also consider the possibility of the player performingadaptive stabilizer circuits. Under adaptive circuits, measurement outcomes on part of the system determine which subsequent operatio...

  53. [53]

    This generalizes the stabilizer measurements, corresponding toµ=0

    n-qudits minimum error success probability via an SDP Here we consider the problem of determining the minimum error success probability of discriminating correctly between 2 states ofnqudits,ρ 0 andρ 1, given that we allow the measurements to have a maximum value of non- stabilizernessµ∈[0,µ max ]. This generalizes the stabilizer measurements, correspondi...

  54. [54]

    We now confirm a known result that projecting into a subspace of a pure state with a pure state leaves a pure state up to a constantα∈[0, 1]in the remaining system

    Proof of Result 1 Before beginning the proof, we re-derive some known proofs from the literature used in the proof of Result 1 for completeness. We now confirm a known result that projecting into a subspace of a pure state with a pure state leaves a pure state up to a constantα∈[0, 1]in the remaining system. Lemma 3.Consider a bipartite spaceH A ⊗ HB, and...

  55. [55]

    Lemma 1P stab suc (⃗∆) = 1 2 + 1 4 ||⃗∆||∞,where|| ⃗∆||∞ is the vector infinity norm i.e., the largest component of the absolute value of ⃗∆

    Proof of Lemma 1 Letρ,σ∈ D(H 1)and ⃗∆ :=n ρ −n σ wheren ρ andn σ are the Bloch vectors ofρandσrespectively. Lemma 1P stab suc (⃗∆) = 1 2 + 1 4 ||⃗∆||∞,where|| ⃗∆||∞ is the vector infinity norm i.e., the largest component of the absolute value of ⃗∆. Proof.When restricted to qubit stabilizer circuits, only a measurement in the eigenbasis ofZfollowing a pre...

  56. [56]

    Proof of Result 2 Firstly, we prove a Lemma for the minimum error success probability under fixed measurements when ρ,σ∈ D(H2), such that ∆ :=ρ−σ= ∑ k ∆kPk,P k ∈ P2 ∀i,j. (H31) In this case, the figure of merit we are considering is Pstab,f suc (∆) = 1 2 + 1 4 max U∈C2 ∑ a∈{0,1}1 ∑ b∈{0,1}1 tr (U†|ab⟩⟨ab|U)∆ , (H32) We note that for two qubits, a stabiliz...

  57. [57]

    Proof of Qubit Helstrom Bound Result 7(Qubit Helstrom Bound [1]).Letρ,σ∈ D(H 1). The optimal minimum error success probability when discrimi- nating betweenρandσis given by P∗ suc(⃗∆) = 1 2 + 1 4 ||⃗∆||2, (H71) where ⃗∆=n ρ −n σ, with nρ,n σ the Bloch vectors ofρandσrespectively, and ||⃗v||2 = q |v1|2 +|v 2|2 +|v 3|2, (H72) is the vector l-2norm. Proof.In...

  58. [58]

    Proof of Result 3 As above in Appendix H 3, we first prove a Lemma for the minimum error success probability under adaptive measurements whenρ,σ∈ D(H2). In this case, the figures of merit is Pstab,a suc (∆) = 1 2 + 1 4 max U∈C2 max {Vb:Vb∈C1}b ∑ a∈{0,1}1 ∑ b∈{0,1}1 tr (U†(V† b |a⟩⟨a|Vb ⊗ |b⟩⟨b|)U)∆ , (H76) where forP stab,a suc (∆), after a global Cliffor...

  59. [59]

    Proof of Result 4 Result. 4 For measurements{M µ 0 ,M µ 1 }bounded byM(M µ i )≤µ∀i, the success probability is tightly bounded below by: Pµ suc ≥max A PA suc(µ), (H112) where PA suc(µ)denotes the optimal success probability assuming that the closest stabilizer state has active support on the sub- spaceA: PA suc(µ) =    1 2 + 1 4N (S+C N||⃗∆||A,1),if ...

  60. [60]

    5 Let us split Alice’s bitstrings in two disjoint sets:X (j) 0 ={x∈ {0, 1}n|xj =0}andX (j) 1 ={x∈ {0, 1}n|xj = 1}

    Proof of Result 5 Result. 5 Let us split Alice’s bitstrings in two disjoint sets:X (j) 0 ={x∈ {0, 1}n|xj =0}andX (j) 1 ={x∈ {0, 1}n|xj = 1}. When restricted to only stabilizer measurements the maximum value possible of Pg, denoted PSTAB g , is given by PSTAB g = 1 2 + 1 4n ∑ j∈[n] ∥⃗∆(j) ∥∞. (H137) where∆ (j) =σ (j) 0 −σ (j) 1 = 1 2⃗∆(j) ·⃗σwith ⃗σa vecto...

  61. [61]

    Then F(U,V)≤P stab,a suc (∆V,|H⟩⟨H|)(H145) where∆ V =V †(|0⟩⟨0| − |1⟩⟨1|)V

    Proof of Result 6 Result6 Let V,U be qubit unitaries where U=C 1TC2 :C 1,C 2 ∈ C1. Then F(U,V)≤P stab,a suc (∆V,|H⟩⟨H|)(H145) where∆ V =V †(|0⟩⟨0| − |1⟩⟨1|)V. Proof.Firstly, let UV † = w00 w01 w10 w11 . (H146) Then F(U,V) = 1 4 |tr UV † |2, (H147) = 1 4 |w00 +w 11|2 (H148) ≤ 1 4 |w00|+|w 11| 2 (H149) = 1 4 |w00|2 +|w 11|2 +2|w 00| |w11| (H150) ≤ 1 2 |w00|...

  62. [62]

    The following Lemma allows the success probability for perform QSD under the optimal coarse graining to be calculated

    Proof of Lemma 2 Consider performing QSD on the statesρ,σ∈ D(C d)using the POVM{M i}N i=1, such thatM i ≥0∀iand ∑N i=1 Mi =I. The following Lemma allows the success probability for perform QSD under the optimal coarse graining to be calculated. Lemma2 If using a POVM{M i}N i=1 for perform QSD, the maximum success probability under all possible coarse grai...