Pith. sign in

REVIEW 4 major objections 5 minor 43 references

Singular value transformation for unknown quantum channels

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read The singular values of a black-box quantum channel can be transformed by QSVT without knowing Kraus operators or preparing purifications.

desk verdict A genuinely new and likely correct block-encoding construction for unknown channels, but with a load-bearing unproven lemma and inconsistent advertised bounds; worth refereeing, not desk-rejecting. read the letter →

arxiv 2506.24112 v2 pith:52ATCIY5 submitted 2025-06-30 quant-ph

Ryotaro Niwa , Zane Marius Rossi , Philip Taranto , Mio Murao This is my paper · ORCID
classification quant-ph
keywords quantumsingularvaluetransformationblock-encodingLiouvillerepresentationchanneldensitymatrixexponentiationmomentsquerycomplexityentanglement-breaking
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 claims that the spectrum of an unknown quantum channel, a non-Hermitian superoperator, can be made accessible to quantum singular value transformation (QSVT) purely from black-box uses of the channel. The authors build an approximate unitary block-encoding of the Hermitized Liouville representation of the channel, with query cost that scales polynomially in the system dimension and inversely in the target precision. This removes the usual requirement of knowing Kraus operators or preparing purifications: once the block-encoding is in hand, any polynomial function of the channel's singular values can be applied by QSVT. The paper demonstrates the payoff on real-order singular value moments, which are relevant for detecting entanglement-breaking channels.

What carries the argument

The central object is the Hermitized Liouville matrix $H = (2d^{1-k})^{-1} \left(\begin{smallmatrix} 0 & E_A \\ E_A^\dagger & 0 \end{smallmatrix}\right)$, whose singular values are those of the channel. The mechanism is a variant of density matrix exponentiation: from the channel's Choi state one builds two states $\rho_\pm$ whose partial transpose over the ancilla is, up to SWAP conjugation, the Hermitized Liouville matrix; applying the interaction $\widetilde H = F_{X\cup Y}\otimes d|\Phi_d^+\rangle\langle\Phi_d^+|_Z$ and tracing out copies of $\rho_\pm$ produces unitaries $e^{\mp i\rho_\pm^{T_Z}\Delta t}$ whose product approximates $e^{-iH}$ with per-step error $O(d\Delta t^2)$. Repeating the steps, conjugating by controlled-SWAPs, and applying an arcsin polynomial through QSVT converts this into the desired block-encoding of $(2/\pi)H$, after which any QSVT polynomial can act on the channel's singular values.

What would settle it

Take a small non-unital channel, such as qubit amplitude damping, classically compute the exact trace-out map defined by Eq. (12) and compare it in diamond norm with the ideal unitary channel generated by $\rho_\pm^{T_Z}$ for a range of step sizes $\Delta t$; if the leading error scales as $d^2\Delta t^2$ rather than $d\Delta t^2$, the query complexity claimed in Theorem 1 must be revised.

Watch

Extended reading notes

Core claim

The paper's central claim is Theorem 1: given black-box access to an unknown channel $\mathcal{E}$ on a $d$-dimensional system, one can construct a quantum channel $\widetilde{U}$ that is $\delta$-close in diamond norm to a $(1,3,0)$-block-encoding unitary of $(2/\pi)H$, where $H = (2d^{1-k})^{-1} \begin{pmatrix} 0 & E_A \\ E_A^\dagger & 0 \end{pmatrix}$ is the Hermitized Liouville representation, using $O(d^{2k+1}\delta^{-1}\log^2(1/\delta))$ queries to $\mathcal{E}$, with $0\le k\le 1/2$ for general channels and $k\le 1$ for unital channels. Because the resulting Hermitian matrix can be fed into QSVT, the singular values of the unknown channel become operands for arbitrary polynomial transformations. The paper also proves an $\Omega(d t^2/\delta)$ lower bound for simulating the corresponding unitary evolution, and uses the block-encoding to estimate $q$-th singular value moments for arbitrary real $q>2$, yielding a direct test for entanglement-breaking channels.

Load-bearing premise

The load-bearing premise is the density-matrix-exponentiation identity in Eq. (12): each short interaction step changes the target state as if the partially transposed copy $\rho_\pm^{T_Z}$ generated the unitary, with error $O(d\Delta t^2)$; if the true error is larger, the total query count $O(d^{2k+1}/\delta)$ no longer follows.

Editorial extensions

If this is right

  • Any polynomial function of the singular values of an unknown channel becomes directly implementable by QSVT, so powers, moments, and filters can be evaluated without tomography or classical postprocessing.
  • Real-order singular value moments $S_q$ for $q>2$, $q\in\mathbb{R}$, can be estimated with the stated query complexity, including the previously inaccessible non-even-integer regime, giving an advantage over Choi-state SWAP circuits for $q>2+2/3$.
  • The first moment $\|R(E_B)\|_1$ can be estimated with $\widetilde O(d^5\|R(E_B)\|_1\log(1/\delta)/\epsilon^3)$ queries to the channel, providing a direct method to test whether a channel is entanglement breaking.
  • The lower bound $\Omega(d t^2/\delta)$ for producing the channel evolution $U_{E_A}(t)$ shows that any black-box protocol must pay a dimension-dependent cost, although the proven upper and lower bounds do not match.

Reading between the lines

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

  • If the error bound in Eq. (12) survives closer scrutiny, the same trace-out construction should block-encode Hermitizations of non-Hermitian linear maps beyond quantum channels, without relying on complete positivity.
  • The paper's own hint about quantum eigenvalue processing suggests that the complex eigenvalues of $E_A$, not just its singular values, may eventually be transformed; that step is not proven here.
  • A useful test is whether the Appendix C reshuffling variant, which estimates the full singular-value spectrum of unital channels with $O(d^6/\epsilon^2)$ samples, also works for near-unital noisy channels in practice.
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

4 major / 5 minor

Summary. This paper considers the task of transforming the singular values of an unknown quantum channel acting on a d-dimensional system, under the assumption that the channel is available only as a black box. The main construction prepares certain states ρ± by applying controlled-SWAP operations to the Choi state of the channel, then uses a 'variant of density matrix exponentiation' (Eq. 12) to approximate unitaries exp(±iρ±^{T_Z}Δt). Combining these gives an approximate implementation of e^{-iH} for a Hermitized Liouville representation H, and QSVT converts this into an approximate (1,3,0)-block-encoding of (2/π)H. Theorem 1 claims a query complexity of O(d^{2k+1}/δ log²(1/δ)) with k∈[0,1/2] for general channels, while the abstract and conclusion state O(d³/δ). Theorem 2 provides a lower bound Ω(dt²/δ) for approximating a particular unitary evolution, and the paper further claims applications to estimating q-th singular value moments for real q>2 and to testing whether a channel is entanglement breaking, including a claimed exponential advantage over Choi-state-based methods.

Significance. If the construction and complexity claims are correct, the paper provides a genuinely new capability: black-box manipulation of channel singular values via QSVT without access to the channel's Kraus operators or a purification of its Choi state. The use of density matrix exponentiation on a non-positive operator (the partial transpose of ρ±) is conceptually interesting and could open the door to further superoperator algorithms. The paper also contains a lower-bound argument based on channel discrimination, which is a valuable complement to the upper bound. However, the significance is currently undermined by three load-bearing problems: the central DME identity is asserted without proof or precise hypotheses; the stated upper bounds are internally inconsistent (abstract/conclusion vs. Theorem 1); and the moment-estimation complexity in Eq. (20) appears to be algebraically incorrect, making the exponential-advantage claim in the applications section unsupported.

major comments (4)
  1. [Theorem 1 proof, Eq. (12)] Equation (12) is the single gate through which every query-complexity claim in the paper passes, yet it is only asserted as a 'variant of density matrix exponentiation [15,20]' without stating the theorem, its hypotheses, or the error constant. The object ρ±^{T_Z} is in general not a valid quantum state, and the interaction Hamiltonian H̃ = F_{X∪Y} ⊗ d|Φ⁺⟩⟨Φ⁺|_Z has operator norm O(d), so the claimed error O(dΔt²) is not an immediate consequence of the standard DME results. The proof must either reproduce the expansion or quote a precise theorem from the literature that applies to the present non-positive partial transpose, with an explicit dimension-dependent error bound. Because the choices Δt = 2δ/d^{k+1} and N = O(d^{2k+1}/δ) are chosen so that N·O(dΔt²)=O(δ), any additional d-dependence in the error would change the advertised query complexity and break Theorem 1. This is the load-bearing wall of the construction, not a peripheral technicality.
  2. [Abstract, Conclusion, and Theorem 1] There is a direct inconsistency between the upper bound stated in the abstract and conclusion, O(d³/δ), and the bound in Theorem 1, O(d^{2k+1}/δ log²(1/δ)) with k∈[0,1/2] for general channels. Since 2k+1 ≤ 2 for k ≤ 1/2, the theorem as stated gives at most O(d²/δ) for general channels, not O(d³/δ). The same issue appears in the lower-bound comparison section, where it is claimed that Theorem 1 implies an eO(d³t²/δ) upper bound for simulating U(t); this does not follow from O(d^{2k+1}/δ) with the stated range of k. The authors should correct either the theorem, the abstract, or the ranges of k, and ensure that all complexity statements in the main text and conclusion are consistent.
  3. [Eq. (20) and Appendix E] The reported query complexity for estimating singular value moments is not consistent with the derivation in Appendix E. From Eq. (E11) with k=1/2, the d-dependence is O(d² · (d^{(q-2)/2}/ε)^{3+2/(q-2)}) = O(d^{3+1.5(q-2)}/ε^{...}) for the unnormalized moment, and after converting to S_q one obtains at least d^{1.5q} scaling, which grows with d. In contrast, Eq. (20) and Eq. (E12) report eO(d log(1/δ)/(d^{1.5(q-2)} ε^{3+2/(q-2)})), which is smaller than 1 for q > 2+2/3 and is impossible for a query complexity. The displayed expression also makes the claimed 'exponential advantage' over SWAP circuits and classical shadow tomography vacuous, since the actual corrected scaling appears to be polynomial growing with d. The authors must re-derive the d-dependence and either substantiate or retract the advantage claim.
  4. [Theorem 3 and Appendix B (samplizer)] Theorem 3 is stated as a 'direct consequence of Thm. 1 and the notion of a samplizer', but no derivation is provided, and the direction of the samplizer result is not obviously applicable. The samplizer theorem quoted in Appendix B converts sample access to a density matrix ρ into a circuit querying a unitary block-encoding of ρ; the present setting has black-box access to a channel E, not copies of a density matrix. The main text does not explain how the samplizer is used to turn the approximate channel construction of Theorem 1 into the QSVT circuit with the stated query count. Since Theorem 3 underlies the moment-estimation and entanglement-breaking applications, this gap should be closed with an explicit proof or a precise reduction.
minor comments (5)
  1. [Eq. (10) and (11)] The circuit diagram in Eq. (10) is not self-contained: the states ρ± are defined only in Eq. (11), but the diagram labels the controlled-SWAP operation on the Choi state without describing the normalizations. Please make the figure caption or the surrounding text clearer.
  2. [Eq. (16)] The displayed equation 'sin H = (⟨+|⊗I)(c-U)(Y⊗I)(c-U†)(|+⟩⊗I)' mixes operator and block-encoding notation; it would be clearer to state explicitly that this yields a (1,1,0)-block-encoding of sin H with the ancilla in the |+⟩ state.
  3. [Appendix F and Appendix H] The comparison of sample complexities in the main text and Appendix H is not uniform: the main text says O(d⁶ log(1/δ)/ε²) for full tomography, while Appendix H states O(d⁶/ε²) without the logarithmic factor. Please align the statements.
  4. [References] Reference [6] contains a malformed arXiv identifier 'arXiv:2303.182241404.6025'; it should be corrected to the actual identifier of Ref. [6].
  5. [Appendix A, Lemma 3 proof] The proof of the upper bound ∥EA∥∞ ≤ √d uses a variational characterization over density matrices; the step stating equality with the spectral norm of EA is terse and would benefit from a one-line justification, since EA is not Hermitian.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the construction derives from external DME and QSVT results; the sole self-citation is a non-load-bearing review.

full rationale

The derivation chain is self-contained: Theorem 1 builds the approximate block-encoding from black-box applications of E via the density-matrix-exponentiation identity in Eq. (12), which is explicitly attributed to Refs. [15,20] (with no author overlap with the present paper), and via standard QSVT/block-encoding lemmas attributed to Gilyén et al. [1]. No parameter is fitted to data and then renamed a prediction; the query complexity O(d^{2k+1}/δ) is obtained by choosing Δt = 2δ/d^{k+1} and summing the stated O(dΔt²) DME error, so the advertised scaling stands or falls with that external error bound rather than with any circular reduction. The only self-citation, Ref. [29] (Martyn, Rossi, Tan, Chuang), appears in Appendix B as a pedagogical pointer, while the load-bearing QSVT theorem is cited to Ref. [1]; hence it is not load-bearing. The unproven status of Eq. (12)'s dimension-dependent error term is a genuine rigor/correctness caveat, not a circularity: the paper does not define the desired result into its assumptions or import a uniqueness claim from its own authors.

Assumptions & free parameters 1 free parameters · 5 assumptions · 0 invented entities

The algorithm introduces no new physical entities; it manipulates the Hermitized Liouville representation, which is a mathematical matrix, not a new force or particle. The main non-standard input is the partial-transpose density matrix exponentiation identity (Eq. 12), which is asserted with an error bound and functions as an unproven axiom in the construction. All other ingredients (QSVT, the samplizer, Chebyshev-based polynomial approximation) are imported from cited prior work. The parameter k in the normalization of H is a tunable design choice rather than a data-fitted constant.

free parameters (1)
  • k = k ∈ [0,1/2] general, k ∈ [0,1] unital; effectively k=1/2 for general channels in the stated O(d^2/δ) complexity
    Chosen by hand to guarantee ||H||∞ ≤ 1/2 while minimizing normalization; it is not fitted to data but the query complexity bound depends on it.
assumptions (5)
  • domain assumption DME partial-transpose identity (Eq. 12): Tr_ρ[e^{-iH̃Δt}(ρ⊗σ)e^{iH̃Δt}] = σ - iΔt[ρ^{T_Z},σ] + O(dΔt^2)
    Used in Theorem 1 to implement e^{-iρ±^{T_Z}Δt}; not fully proven in the main text and is load-bearing for the query complexity.
  • standard math QSVT polynomial approximation: there exists an efficiently computable odd polynomial P of degree O(log(1/δ)) with |P(x)-(2/π)arcsin x| ≤ δ on [-1/2,1/2]
    Quoted from Gilyén et al. [1, Lemma 70]; standard.
  • standard math Samplizer theorem (Thm. 5 in Appendix B): converts unitary-query circuits into state-copy circuits with O(Q^2/δ log^2(Q/δ)) samples
    Quoted from Wang & Zhang [23]; used for moment estimation.
  • standard math Polynomial approximation of x^{q-2} with degree O(1/ε^{1/(q-2)}) uniformly on [-1,1]
    Based on averaging Chebyshev polynomials as in Liu & Wang [12]; used for S_q with real q.
  • domain assumption Channel discrimination lower bound: teleportation-covariant channels reduce adaptive discrimination to Choi-state trace distance [37]
    Used in the lower bound proof (Appendix D).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Singular value transformation for unknown quantum channels." pith.science (2026). https://pith.science/paper/52ATCIY5

@misc{pith2026250624112,
  author       = {Pith},
  title        = {Pith review of: Singular value transformation for unknown quantum channels},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/52ATCIY5}},
  note         = {Machine review of arXiv:2506.24112}
}
abstract

Given the ability to apply an unknown quantum channel acting on a $d$-dimensional system, we develop a quantum algorithm for transforming its singular values. The spectrum of a quantum channel as a superoperator is naturally tied to its Liouville representation, which is in general non-Hermitian. Our key contribution is an approximate block-encoding scheme for this representation in a Hermitized form, given only black-box access to the channel; this immediately allows us to apply polynomial transformations to the channel's singular values by quantum singular value transformation (QSVT). We then demonstrate an $O(d^3/\delta)$ upper bound and an $\Omega(d/\delta)$ lower bound for the query complexity of constructing a quantum channel that is $\delta$-close in diamond norm to a block-encoding of the Hermitized Liouville representation. We show our method applies practically to the problem of learning the $q$-th singular value moments of unknown quantum channels for arbitrary $q>2, q\in \mathbb{R}$, which has implications for testing if a quantum channel is entanglement breaking.

Figures

Figures reproduced from arXiv: 2506.24112 by the authors.

Figure 1
Figure 1. FIG. 1. (a) Approximate block-encoding scheme for the Li [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. A quantum circuit for calculating the 4-th singular value moment. [PITH_FULL_IMAGE:figures/full_fig_p018_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 26 canonical work pages

  1. [15]

    Chen and L.-A

    K. Chen and L.-A. Wu, A matrix realignment method for recognizing entanglement, Quantum Inf. Comput. 3, 193 (2003), arXiv:quant-ph/0205017 [quant-ph]

    arXiv 2003

  2. [1]

    Gily´ en, Y

    A. Gily´ en, Y. Su, G. H. Low, and N. Wiebe, Quan- tum singular value transformation and beyond: expo- nential improvements for quantum matrix arithmetics, in Proc. 51st Annu. ACM SIGACT Symp. Theory Comput. (2019) p. 193, arXiv:1806.01838 [quant-ph]

    arXiv 2019

  3. [2]

    The states ρ± have the form ρ± = 1 2 |0⟩⟨0| ⊗(E ⊗I)|Φ+ d ⟩⟨Φ+ d | ±1 2d |0⟩⟨1| ⊗(EAFY Z)TZ ± 1 2d |1⟩⟨0| ⊗(FY ZE † A)TZ + 1 2 |1⟩⟨1| ⊗(I ⊗ E)|Φ+ d ⟩⟨Φ+ d |, (11) where FY Zdenotes the SWAP operator swapping HY and HZ, and ( ·)TZ denotes the partial transposition with re- spect to HZ. Utilizing a variant of density matrix exponentia- tion [15, 20], the uni...

    work page

  4. [3]

    G. H. Low, T. J. Yoder, and I. L. Chuang, Methodology of Resonant Equiangular Composite Quantum Gates, Phys. Rev. X 6, 041067 (2016), arXiv:1603.03996 [quant-ph]

    arXiv 2016

  5. [4]

    Camps, L

    D. Camps, L. Lin, R. V. Beeumen, and C. Yang, Explicit Quantum Circuits for Block Encodings of Certain Sparse Matrices (2023), arXiv:2203.10236 [quant-ph]

    arXiv 2023

  6. [5]

    S¨ underhauf, E

    C. S¨ underhauf, E. Campbell, and J. Camps, Block- encoding structured matrices for data input in quantum computing, Quantum 8, 1226 (2024), arXiv:2302.10949 [quant-ph]

    arXiv 2024

  7. [7]

    J. J. Wallman and S. T. Flammia, Randomized bench- marking with confidence, New J. Phys. 16, 103032 (2014), arXiv:2303.182241404.6025 [quant-ph]

    arXiv 2014

  8. [8]

    Lindblad, On the generators of quantum dynamical semigroups, Commun

    G. Lindblad, On the generators of quantum dynamical semigroups, Commun. Math. Phys. 48, 119 (1976)

    1976

Show all 43 references
  1. [9]

    Jamio lkowski, Linear transformations which preserve trace and positive semidefiniteness of operators, Rep

    A. Jamio lkowski, Linear transformations which preserve trace and positive semidefiniteness of operators, Rep. Math. Phys. 3, 275 (1972)

    1972

  2. [10]

    Choi, Completely positive linear maps on complex matrices, Linear Algebra Appl

    M.-D. Choi, Completely positive linear maps on complex matrices, Linear Algebra Appl. 10, 285 (1975)

    1975

  3. [11]

    Cai and W

    J. Cai and W. Song, Novel Schemes for Directly Mea- suring Entanglement of General States, Phys. Rev. Lett. 101, 190503 (2008), arXiv:0804.2246 [quant-ph]

    2008 arXiv

  4. [12]

    Z. Liu, Y. Tang, H. Dai, P. Liu, S. Chen, and X. Ma, De- tecting Entanglement in Quantum Many-Body Systems via Permutation Moments, Phys. Rev. Lett. 129, 260501 (2022), arXiv:2203.08391 [quant-ph]

    2022 arXiv

  5. [13]

    Liu and Q

    Y. Liu and Q. Wang, On Estimating the Trace of Quantum State Powers, in Proc. 2025 Annu. ACM-SIAM Symp. Discrete Algorithms (2025) p. 947, arXiv:2410.13559 [quant-ph]

    2025 arXiv

  6. [14]

    Rudolph, Further Results on the Cross Norm Criterion for Separability, Quantum Inf

    O. Rudolph, Further Results on the Cross Norm Criterion for Separability, Quantum Inf. Process. 4, 219 (2005), arXiv:quant-ph/0202121 [quant-ph]

    2005 arXiv

  7. [16]

    F. Wei, Z. Liu, G. Liu, Z. Han, D.-L. Deng, and Z. Liu, Simulating non-completely positive actions via exponen- tiation of Hermitian-preserving maps, npj Quantum Inf. 10, 134 (2024), arXiv:2308.07956 [quant-ph]

    2024 arXiv

  8. [17]

    The Liouville representation EA and the Choi stateEB are sometimes called the A-form and the B-form matrices, respectively (as they were originally derived in Ref. [27])

  9. [18]

    We note that the (unnormalized) Choi matrix dEB is also often considered in the literature, which we distinguish 6 in this manuscript from the Choi state EB

  10. [19]

    Chiribella, G

    G. Chiribella, G. M. D’Ariano, and P. Perinotti, Theo- retical framework for quantum networks, Phys. Rev. A 80, 022339 (2009), arXiv:0904.4483 [quant-ph]

    2009 arXiv

  11. [20]

    S. Milz, F. A. Pollock, and K. Modi, An Introduction to Operational Quantum Dynamics, Open Syst. Inf. Dyn. 24, 1740016 (2017), arXiv:1708.00769 [quant-ph]

    2017 arXiv

  12. [21]

    Lloyd, M

    S. Lloyd, M. Mohseni, and P. Rebentrost, Quantum prin- cipal component analysis, Nat. Phys. 10, 631 (2014), arXiv:1307.0401 [quant-ph]

    2014 arXiv

  13. [22]

    Kimmel, C

    S. Kimmel, C. Y.-Y. Lin, G. H. Low, M. Ozols, and T. J. Yoder, Hamiltonian simulation with optimal sample com- plexity, npj Quantum Inf. 3, 1 (2017), arXiv:1608.00281 [quant-ph]

    2017 arXiv

  14. [23]

    G. H. Low and I. L. Chuang, Hamiltonian Simulation by Qubitization, Quantum 3, 163 (2019), arXiv:1610.06546 [quant-ph]

    2019 arXiv

  15. [24]

    Wang and Z

    Q. Wang and Z. Zhang, Time-Efficient Quantum Entropy Estimator via Samplizer, in 32nd Annual European Sym- posium on Algorithms (ESA 2024), Leibniz International Proceedings in Informatics (LIPIcs), Vol. 308, edited by T. Chan, J. Fischer, J. Iacono, and G. Herman (Schloss Dag...

    2024

  16. [25]

    Horodecki, P

    M. Horodecki, P. W. Shor, and M. B. Ruskai, Entan- glement Breaking Channels, Rev. Math. Phys. 15, 629 (2003), arXiv:quant-ph/0302031 [quant-ph]

    2003 arXiv

  17. [26]

    Knill and R

    E. Knill and R. Laflamme, Power of One Bit of Quan- tum Information, Phys. Rev. Lett. 81, 5672 (1998), arXiv:quant-ph/9802037 [quant-ph]

    1998 arXiv

  18. [27]

    G. H. Low and Y. Su, Quantum Eigenvalue Processing, in 2024 IEEE 65th Annu. Symp. Found. Comput. Sci. (2024) p. 1051, arXiv:2401.06240 [quant-ph]

    2024

  19. [28]

    E. C. G. Sudarshan, P. M. Mathews, and J. Rau, Stochas- tic Dynamics of Quantum-Mechanical Systems, Phys. Rev. 121, 920 (1961)

    1961

  20. [29]

    Watrous, The Theory of Quantum Information(Cam- bridge Univ

    J. Watrous, The Theory of Quantum Information(Cam- bridge Univ. Press, Cambridge, 2018)

    2018

  21. [30]

    J. M. Martyn, Z. M. Rossi, A. K. Tan, and I. L. Chuang, Grand Unification of Quantum Algorithms, PRX Quan- tum 2, 040203 (2021), arXiv:2105.02859 [quant-ph]

    2021 arXiv

  22. [31]

    G. H. Low and I. L. Chuang, Optimal Hamiltonian Sim- ulation by Quantum Signal Processing, Phys. Rev. Lett. 118, 010501 (2017), arXiv:1606.02685 [quant-ph]

    2017 arXiv

  23. [32]

    Tang and K

    E. Tang and K. Tian, A CS guide to the quantum singular value transformation (2023), arXiv:2302.14324 [quant- ph]

    2023 arXiv

  24. [33]

    In truth we have to know milder things, like a bound on the range of possible singular values, and the span of the left and right singular vectors respectively, but these are often trivially known

  25. [34]

    Keyl and R

    M. Keyl and R. F. Werner, Estimating the spectrum of a density operator, Phys. Rev. A 64, 052311 (2001), arXiv:quant-ph/0102027 [quant-ph]

    2001 arXiv

  26. [35]

    O’Donnell and J

    R. O’Donnell and J. Wright, Efficient quantum tomogra- phy II (2016), arXiv:1612.00034 [quant-ph]

    2016 arXiv

  27. [36]

    Chiribella, G

    G. Chiribella, G. M. D’Ariano, and P. Perinotti, Trans- forming quantum operations: Quantum supermaps, EPL (Europhys. Lett.) 83, 30004 (2008), arXiv:0804.0180 [quant-ph]

    2008 arXiv

  28. [37]

    Gour, Comparison of Quantum Channels by Super- channels, IEEE Trans

    G. Gour, Comparison of Quantum Channels by Super- channels, IEEE Trans. Inf. Theory 65, 5880 (2019), arXiv:1808.02607 [quant-ph]

    2019 arXiv

  29. [38]

    Pirandola and C

    S. Pirandola and C. Lupo, Ultimate precision of adaptive noise estimation, Phys. Rev. Lett. 118, 100502 (2017), arXiv:1609.02160 [quant-ph]

    2017 arXiv

  30. [39]

    7 SUPPLEMENT AL MA TERIAL DIAGRAM NOT A TIONS Here we introduce diagram notations that facilitate calculations

    Ud consists of Heisenberg-Weyl operators, which is equiv- alent to Pauli operators for n qubit systems. 7 SUPPLEMENT AL MA TERIAL DIAGRAM NOT A TIONS Here we introduce diagram notations that facilitate calculations. We assign an index to each bra and ket degrees of freedom, an...

  31. [40]

    Also, right eigenoperator is traceless except for the fixed point

    If λn is an eigenvalue, λ∗ n is also an eigenvalue. Also, right eigenoperator is traceless except for the fixed point

  32. [41]

    |λn| ≤1. Proof. 1; The following proof is given in [28]. Define a CPTP map Φ as Φn(ρ) = 1 2n 2n−1X k=0 E (k)(ρ). (A5) We define the following set Dn ≡ {Φn(ρ)|ρ ∈ L(H)}. It holds that Φn+1(ρ) = 1 2n+1 2n+1−1X k=0 E (k)(ρ) = 1 2 Φn(ρ) + 1 2 Φn(E (n)(ρ)) = Φn 1 2 ρ + 1 2 E (n)(ρ)...

  33. [42]

    The Hadamard test The Hadamard test calculates Tr( ρU ) by repeatedly running the following circuit. |0⟩ H H ρ U More specifically, the binary measurement outcome obeys the probability distribution p(0) = 1 + Re[Tr(ρU )] 2 p(1) = 1 − Re[Tr(ρU )] 2 (B1) so if we record +1 when ...

  34. [43]

    Quantum singular value transformation (QSVT) The quantum singular value transformation (QSVT) is a framework that encompasses many well-known algorithms such as amplitude amplification, matrix inversion, and Hamiltonian simulation [1, 29]. Specifically, one can show that the c...

  35. [44]

    This method converts a block-encoding unitary access model into a quantum state access model

    Samplizer Here we review the notion of the samplizer, which generalizes the density matrix exponentiation trick [20], originally introduced in [23]. This method converts a block-encoding unitary access model into a quantum state access model. Theorem 5. (Samplizer) Suppose C =...

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.