Pith. sign in

REVIEW 3 major objections 6 minor 1 cited by

Efficient Quantum Access Model for Sparse Structured Matrices using Linear Combination of Things

T0 review · 3 major / 6 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 A five-operator Sigma basis makes LCU decompositions of sparse PDE matrices polylogarithmic in size.

desk verdict A useful circuit-construction paper with a correct core, but the polylog term-count headline is conditional on a pattern-discovery step the paper admits it cannot automate. read the letter →

arxiv 2507.03714 v2 pith:ZPODVGCC submitted 2025-07-04 quant-ph cs.NAmath.NA

Abeynaya Gnanasekaran , Amit Surana This is my paper · ORCID
classification quant-phcs.NAmath.NA MSC 81P68
keywords SigmabasisLCNUunitarycompletionsparsestructuredmatricesPDEdiscretizationblockencodingvariationalquantumalgorithmsmulti-controlledToffoli
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 replacing the Pauli basis in Linear Combination of Unitaries (LCU) decompositions with a five-element set of simple non-unitary operators, the Sigma basis, can cut the number of terms needed to represent sparse structured matrices from polynomial down to polylogarithmic scaling in the matrix size. This matters because the number of terms drives the measurement cost in variational quantum algorithms and the gate cost of block encodings in fault-tolerant algorithms. The authors show that matrices arising from 1D Poisson, heat, and wave discretizations decompose into O(log N) Sigma-basis terms, and that each non-unitary term can be embedded into a unitary circuit using at most n+1 single-qubit gates and one multi-controlled Toffoli gate. They also give numerical and semi-analytical recipes for finding such decompositions and demonstrate exponential reductions in term count relative to Pauli decompositions on those PDE examples.

What carries the argument

The load-bearing object is the Sigma basis S = {I, σ+ = |0⟩⟨1|, σ− = |1⟩⟨0|, σ+σ− = |0⟩⟨0|, σ−σ+ = |1⟩⟨1|}. A matrix is written as a weighted sum of tensor products of these five factors, each product being non-unitary but binary-valued. The argument then rests on unitary completion: any non-unitary term Al = ⊗k σk is embedded in a larger unitary Ul = [[Al, Ac_l],[Ac_l, Al]] whose circuit can be built without ever computing the complement Ac_l explicitly. Theorem 1 gives the completion operator as the tensor product of the completions of the factors, and Theorem 2 shows the resulting circuit uses only the completion gates plus a single C^mX gate whose controls are fixed by the pattern of σ+/σ− factors. The logarithmic term counts come from a separate mechanism: the PDE matrices are decomposed recursively, as block matrices of the form I2 ⊗ $A^{{(s-1)}}$ + σ− ⊗ $D^{{(s-1)}}$ + σ+ ⊗ ($D^{{(s-1)}}$)^T, with each off-recursion piece $D^{{(s-1)}}$ represented by a single pure-σ+ tensor product, so the recursion depth supplies only logarithmically many terms.

What would settle it

Assemble the recursion (13) for the 1D Poisson matrix with nx = 128, giving 2 log2 nx + 1 = 15 Sigma-basis terms, sum the terms with their coefficients, and subtract from the tridiagonal matrix (12); a nonzero residual anywhere would refute the claimed logarithmic decomposition. A complementary scope test repeats this for the 2D Neumann Poisson matrix and checks whether boundary terms force the count above polylog(N), as the paper's Θ($n^{{d-1/d}}$) analysis predicts.

Watch

Extended reading notes

Core claim

For matrices that arise from standard finite-difference discretizations of 1D Poisson, heat, and wave equations, the paper establishes that a Linear Combination of Non-Unitaries (LCNU) decomposition over the Sigma basis uses only O(log N) terms, where N is the matrix dimension, in contrast to Pauli-basis LCU decompositions whose term counts grow linearly or faster. The concrete counts are 2 log nx + 1 terms for the Poisson matrix, 4 log nx + 6 for the heat-system matrix, and log nt + 1 + 2(2 log 2nx + 4) for the wave-system matrix. Because the Sigma basis elements are non-unitary, each term is handled by unitary completion: Theorem 1 states that the completion of a tensor product is the tensor product of completions, where σ+ and σ− complete to σx and the other three basis elements complete to the identity. Algorithm 1 converts that completion into a circuit with at most n+1 single-qubit gates and one C^mX gate, m ≤ n, and Theorem 2 proves the construction. The same machinery yields Hadamard-test circuits for variational cost functions and block encodings for fault-tolerant algorithms, with resource estimates showing the asymptotic costs match Pauli-based block encoding except that L is exponentially smaller.

Load-bearing premise

The polylogarithmic term count is not a property of the Sigma basis alone: it depends on the matrix possessing a recursive or telescoping block structure with only O(1) nonzeros lying outside the recursion at each level, and the paper concedes there is no general recipe for recognizing that pattern.

Editorial extensions

If this is right

  • For 1D Poisson, heat, and wave discretizations, the number of LCU terms drops from O(N) or worse under the Pauli basis to O(log N), so the quadratic measurement overhead of algorithms like VQLS becomes polylogarithmic.
  • Each Sigma-basis term can be unitarized by a circuit with at most n+1 single-qubit gates and one C^mX gate, and since multi-controlled Toffoli gates admit decompositions of depth O(log m) and size O(m), per-term circuit overhead stays logarithmic in the matrix size.
  • The unitary-completion circuits plug directly into Hadamard tests for ⟨ψ1|Al|ψ2⟩ and ⟨ψ1|Ai* M Aj|ψ2⟩, so both global and local VQA cost functions can be evaluated without bespoke observables.
  • Block encodings of the PDE matrices can be built from the Sigma-basis LCNU decomposition using PREP and SELECT routines, and with L polylogarithmic the gate count, depth, and ancilla count of the block encoding inherit the exponential improvement.
  • Unitary completion is strictly cheaper than unitary dilation for these terms: completion needs one C^mX gate, while the dilation form requires 2s+1 of them, so the construction wins on circuit depth.

Reading between the lines

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

  • The recursion pattern behind the 1D examples is not limited to finite differences: any block-Toeplitz or telescoping matrix whose off-recursion blocks are rank-one tensor products should admit the same logarithmic decomposition, so structured finite-element and multigrid operators are natural next targets.
  • The paper's own scaling analysis points to a sharp boundary: in d dimensions, Neumann or Robin boundaries contribute Θ(nx^{(d-1)/d}) boundary terms, so for these boundary conditions the polylogarithmic advantage cannot survive without additional spatial uniformity of the boundary terms.
  • Because the basis can be freely extended with any easily implementable operator such as Pauli matrices, permutation matrices, or other low-depth unitaries, a practical recipe suggests itself: choose an extended 'things' basis per matrix family, optimizing the trade-off between fewer terms and deeper completion circuits.
  • A testable consequence is that the numerical decomposition method can serve as a diagnostic: for a given sparse matrix, if the minimal Sigma-basis term count grows like nnz(A) rather than polylog(N), the matrix lacks the recursive structure needed for the exponential speedup.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. The paper proposes a framework for decomposing a matrix A into a linear combination of tensor products of the five-element Sigma basis S = {I, sigma_+, sigma_-, sigma_+ sigma_-, sigma_- sigma_+}, with the non-unitary terms handled by unitary completion. It gives Algorithm 1 for constructing the completion circuit for each term, Theorem 2 bounding the circuit cost by n+1 single-qubit gates and one C^mX gate with m <= n, Theorem 3 for single-entry matrices, and Theorem 4 comparing unitary completion with unitary dilation. It then develops numerical and semi-analytical decomposition methods, applies them to 1D Poisson, heat, and wave discretizations to obtain O(log N) terms, and sketches Hadamard-test and block-encoding constructions for variational and fault-tolerant quantum algorithms.

Significance. If the claimed polylogarithmic term counts and low circuit overhead hold, this is a practically useful quantum access model for structured PDE matrices: it would replace O(N)-term Pauli LCUs with O(polylog N)-term Sigma LCUs at modest circuit cost, giving an exponential reduction in LCU measurement overhead. The main strengths are the explicit proofs of Theorems 2 and 3, the reproducible pseudocode in Algorithm 1, the verifiable worked decompositions for the three 1D PDE examples, and the clean comparison with unitary dilation showing that completion uses fewer multi-controlled gates. The contribution is limited by the absence of an automatic pattern-discovery algorithm and by unqualified scaling claims in the abstract and conclusions; both issues are fixable in revision.

major comments (3)
  1. [Abstract, Section I, Section V.4] The abstract and introduction advertise 'decompositions with only polylogarithmic scaling in the number of terms' and 'exponential improvements in decomposition size while retaining circuit efficiency' as properties of the Sigma-basis method. This is stronger than what the body establishes. Section IV states that 'There is no general recipe for determining such patterns and one has to proceed on a case-by-case basis,' and Section V.4 concedes that for Neumann and Robin boundary conditions in d dimensions the boundary terms scale as Theta(n_x^((d-1)/d)), which destroys polylog scaling unless the boundary terms are spatially uniform, and that this 'must be assessed on a case-by-case basis.' The abstract and conclusions should be rephrased to present polylog scaling as a property of a recursively structured subclass, demonstrated on the 1D Poisson/heat/wave examples, rather than as the generic behavior of Sigma-basis decompositions.
  2. [Section IV] The semi-analytical approach is a case-by-case pattern-recognition procedure, not an algorithm. The paper gives recursions for A_1, tilde-I, A_e, and A_p, but no method to discover a recursive or telescoping structure in a new matrix, no decision procedure, and no characterization of when a polylogarithmic decomposition exists. For a general matrix the only guaranteed method is Theorem 3, which produces up to nnz(A) terms and hence no exponential advantage. Since the central scalability claim depends on this pattern-discovery step, the paper should either supply an algorithmic procedure for finding the pattern or explicitly scope its contribution to the class of matrices for which such a pattern is supplied.
  3. [Section VI.2.1] The block-encoding resource estimate is asserted rather than derived. The text states that two n-Toffoli gates and the C^mX gate give C_ctrl(U_l, n+2) = O(n) and D_ctrl(U_l, n+2) = O(log n), citing [29] and [22], but it never shows the explicit Clifford+T decomposition of the controlled-U_l circuit in Figure 7, and Table I is for Pauli-string SELECT/PREP and is not specialized to the Sigma basis. In particular, the claim of 'the same asymptotic gate count ... at the expense of two additional ancilla bits' is an extrapolation. The authors should provide the complete derivation of the SELECT and PREP costs for LCNU terms, including the two extra ancillas and the C^mX construction, or qualify the resource claims as estimates pending that derivation.
minor comments (6)
  1. [Section III.2, Lemma 2] In the proof of Lemma 2, the second and third displayed implications both state 'sigma_0(0,1) = 1'; the second should be 'sigma_0(1,0) = 1'. The reverse direction is left as an exercise; please include it for completeness.
  2. [Figure 5] The caption says the black dotted line is y = x, but for the heat and wave panels the Pauli-term curves grow faster than linear. State in the caption that the y = x line is only a reference for the Poisson panel, or remove it.
  3. [Section VI.2.1] The phrase 'two n-Toffoli gates (one control, n targets)' is ambiguous; specify whether n is the number of controls or the size of the target register, and reconcile it with the C^mX gate count used in the same paragraph.
  4. [Table II] Several truth tables in Table II appear to contain duplicate rows, particularly for inputs with binary representations 110 and 111; please regenerate the tables and check each row against the corresponding circuit.
  5. [Section V.1] The sentence 'The result also extends to d-dimensional Poisson PDE with Dirichlet boundary condition' is not demonstrated in the text. A construction, a precise statement of the conditions, or a citation to a proof should be provided.
  6. [Throughout] There are several typos, including 'Possion' in the Figure 5 caption, 'structued' in Section V, and 'desired desired' in Section VI.1.1.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the polylog term-count results are constructive algebraic identities and the main external load from the authors' prior work is an independent published theorem, not a fitted input.

full rationale

The paper's central derivations are constructive and self-contained. The Poisson decomposition in Eq. (13) is an explicit recursive identity; the heat and wave decompositions in Secs. V.2-V.3 are built from the same verified recursion plus explicit boundary-term identities, and the term counts are read off these expressions rather than fitted to the claimed scaling. The numerical approach in Sec. IV is also an exact identity: each nonzero entry is written as one Sigma-basis tensor product via Theorem 3, so the nnz(A) bound is by construction but not a hidden prediction. The circuit-cost claims in Theorem 2 are proven in the paper with an elementary permutation-matrix argument; the only external load is Theorem 1, whose proof is cited from the authors' prior work [12]. That theorem is a published, parameter-free mathematical statement with explicit assumptions, and it does not invoke the present paper's conclusions, so under the review rules it is real evidence rather than circularity. The paper itself flags the limits of the approach: Sec. IV states 'There is no general recipe for determining such patterns and one has to proceed on a case-by-case basis,' and Sec. V.4 concedes that Neumann/Robin boundary corrections scale as Theta(nx^((d-1)/d)) and may destroy polylog scaling unless spatially uniform. These are scope limitations and algorithmic-completeness caveats, not circular reductions. Similarly, the block-encoding resource estimates in Sec. VI.2.1 inherit constructions from [29] and assert the same asymptotic complexity for the Sigma-basis SELECT; even if that step is an extrapolation rather than a full derivation, it is not a case of a fitted parameter being renamed as a prediction or a conclusion being equivalent to its input by definition. Overall, the paper's predictions are verified against direct decompositions and external Pauli-basis comparisons, so no specific circular step can be exhibited.

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

No free parameters are fitted: the decomposition coefficients derive from the matrix being decomposed, not from data fitting. No new physical entities are introduced; the Sigma basis, unitary completion, and block-encoding machinery are mathematical constructions from prior literature. The central assumptions are that Sigma-basis tensor products are partial isometries (so completion applies) and that the cited C^mX implementation achieves its stated resource bounds.

assumptions (5)
  • standard math Sigma basis spans all 2x2 complex matrices
    Used in Thm 3 and Sec VIII to justify single-entry tensor product decomposition of arbitrary matrices.
  • standard math Every partial isometry has a unitary completion
    Invoked in Definition 2 and Theorem 1 via Ex 2.67 of [13].
  • domain assumption Tensor products of Sigma basis elements are partial isometries
    Stated as Lemma 1 of [12]; required for the unitary completion block form of Ul in Def 3 and Thm 2.
  • domain assumption C^mX gates can be implemented with O(m) size and O(log m) depth using one ancilla
    Depends on the construction of [22]; used for resource estimates in Sec VI.1.3 and VI.2.1.
  • standard math Linear combination of block-encodings preserves block-encoding
    Used in Sec VI.2 to combine the Ul completions into a block-encoding of A; cited from [28].

how reviews work

0 comments
Cite this review

Pith. "Pith review of Efficient Quantum Access Model for Sparse Structured Matrices using Linear Combination of Things." pith.science (2026). https://pith.science/paper/ZPODVGCC

@misc{pith2026250703714,
  author       = {Pith},
  title        = {Pith review of: Efficient Quantum Access Model for Sparse Structured Matrices using Linear Combination of Things},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZPODVGCC}},
  note         = {Machine review of arXiv:2507.03714}
}
read the original abstract

We present a novel framework for Linear Combination of Unitaries (LCU)-style decomposition tailored to structured sparse matrices, which frequently arise in the numerical solution of partial differential equations (PDEs). While LCU is a foundational primitive in both variational and fault-tolerant quantum algorithms, conventional approaches based on the Pauli basis can require a number of terms that scales quadratically with matrix size. We introduce the Sigma basis, a compact set of simple, non-unitary operators that can better capture sparsity and structure, enabling decompositions with only polylogarithmic scaling in the number of terms. We develop both numerical and semi-analytical methods for computing Sigma basis decompositions of arbitrary matrices. Given this new basis is comprised of non-unitary operators, we leverage the concept of unitary completion to design efficient quantum circuits for evaluating observables in variational quantum algorithms and for constructing block encodings in fault-tolerant quantum algorithms. We compare our method to related techniques like unitary dilation, and demonstrate its effectiveness through several PDE examples, showing exponential improvements in decomposition size while retaining circuit efficiency.

Figures

Figures reproduced from arXiv: 2507.03714 by the authors.

Figure 1
Figure 1. FIG. 1: Circuit for [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2: Circuit for [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 4
Figure 4. FIG. 4: Illustration of a recursive/ telescoping matrix structure that leads to a [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
Figures from the paper (3 more)
Figure 5
Figure 5. Figure 5: FIG. 5: Comparison of number of terms in the Sigma basis and Pauli basis decomposition for [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]
Figure 6
Figure 6. Figure 6: FIG. 6: Hadamard test circuits for computing two terms which typically arise in VQAs. [PITH_FULL_IMAGE:figures/full_fig_p015_6.png]
Figure 7
Figure 7. Figure 7: FIG. 7: Circuit for constructing controlled- [PITH_FULL_IMAGE:figures/full_fig_p017_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. A Scalable Approach to Solve the Carleman Linearized Burgers' Equation on a Quantum Computer

    quant-ph 2026-07 conditional novelty 6.5 of 10

    LCNU loading plus multigrid-warmed VQLS solves Carleman-linearized 1D Burgers on quantum hardware/simulators, with circuits scaling to 2^80 points.

Reference graph

Works this paper leans on

43 extracted references · 40 canonical work pages · cited by 1 Pith paper

  1. [12]

    Ef- ficient variational quantum linear solver for structured sparse matrices

    Abeynaya Gnanasekaran and Amit Surana. Ef- ficient variational quantum linear solver for structured sparse matrices. In 2024 IEEE In- ternational Conference on Quantum Comput- ing and Engineering (QCE), volume 1, pages 199–210. IEEE, 2024

    work page 2024

  2. [29]

    Variational quantum solutions to the advection–diffusion equation for applications in fluid dynamics

    Reuben Demirdjian, Daniel Gunlycke, Car- olyn A Reynolds, James D Doyle, and Sergio Tafur. Variational quantum solutions to the advection–diffusion equation for applications in fluid dynamics. Quantum Information Process- ing, 21(9):322, 2022

    work page 2022

  3. [22]

    Predicting many properties of a quan- tum system from very few measurements

    Hsin-Yuan Huang, Richard Kueng, and John Preskill. Predicting many properties of a quan- tum system from very few measurements. Na- ture Physics, 16(10):1050–1057, 2020

    work page 2020

  4. [1]

    Compute the unitary completion of Al as ¯Al = σ− ⊗ ¯I ⊗ σ+σ− = σx ⊗ I ⊗ I

    work page

  5. [2]

    Construct the single-qubit gates X on q0 and I on q1, q2

    work page

  6. [3]

    Thus, we get a total of 1+2(k−1)+1+1 = 2 k+1 C nX gates completing the proof

    Finally, permute rows in position r′ 1 and r1 again using one C nX gate. Thus, we get a total of 1+2(k−1)+1+1 = 2 k+1 C nX gates completing the proof

    work page

  7. [4]

    In this case, k = 1 as the second term in the tensor product is I

    Compute k as the number of factors in representation of Al that are Identity op- erations. In this case, k = 1 as the second term in the tensor product is I

    work page

  8. [5]

    a0 X q0 X q1 I q2 I FIG

    Construct a C 2X gate (CCNOT gate) by adding a closed control onq0, open control on q2 and a X gate on target a0. a0 X q0 X q1 I q2 I FIG. 1: Circuit for Ul corresponding to the term Al = σ− ⊗ I ⊗ σ+σ−. We have successfully constructed the cir- cuit in five simple steps. Readers not inter- ested in the proof of correctness of Algorithm 1 may skip the foll...

    work page

Show all 43 references
  1. [6]

    Note that most significant bit corresponds to the ancilla qubit

    There are two non-zero rows r1, r2 in AlAT l where r1 = {0100} and r2 = {0110}. Note that most significant bit corresponds to the ancilla qubit. The binary representation of these rows differ in the q1 bit as the corresponding factor in Al is I and corresponds to case 2 in the...

  2. [7]

    The second permutation (r2 ↔ r′

    can be done using a C 3X with closed control on q0 and open control on q1, q2. The second permutation (r2 ↔ r′

  3. [8]

    The target is always on the ancillaa0

    can be done with another C 3X with closed control on q0, q1 and open control on q2. The target is always on the ancillaa0. However, as the two C 3X gates differ only in the control operation on q1, they can be effectively repre- sented by a single C 2X gate as shown in Fig. 3....

  4. [9]

    Finally using the procotols defined in Lem 6, 7 and following the same analysis in the proof of Thm

    = O(n), Dctrl(Ul, n+ 2) = O(log n). Finally using the procotols defined in Lem 6, 7 and following the same analysis in the proof of Thm. 4 in [29], we obtain the same asymptotic gate count and circuit depth for Sigma basis decomposition that is summarized in Table I at the exp...

  5. [10]

    This can be done using a single C nX gate by Lemma 1 as they differ by one bit

    Permute rows r1 and r′ 1 = 2 n + r1. This can be done using a single C nX gate by Lemma 1 as they differ by one bit. Now row r1 is in position r′ 1

  6. [11]

    These rows differ by k bits and require 2( k − 1)+1 C nX gates by induction hypothesis

    Permute row in position r′ 1 and r2. These rows differ by k bits and require 2( k − 1)+1 C nX gates by induction hypothesis. Now original row r1 is in position r2 and original row in position r2 is in position r′ 1

  7. [13]

    The variational quan- tum eigensolver: a review of methods and best practices

    Jules Tilly, Hongxiang Chen, Shuxiang Cao, Dario Picozzi, Kanav Setia, Ying Li, Ed- ward Grant, Leonard Wossnig, Ivan Rungger, George H Booth, et al. The variational quan- tum eigensolver: a review of methods and best practices. Physics Reports, 986:1–128, 2022

    2022

  8. [14]

    Variational quantum linear solver

    Carlos Bravo-Prieto, Ryan LaRose, Marco Cerezo, Yigit Subasi, Lukasz Cincio, and Patrick J Coles. Variational quantum linear solver. Quantum, 7:1188, 2023

    2023

  9. [15]

    Qubitization of arbitrary basis quantum chem- istry leveraging sparsity and low rank factor- ization

    Dominic W Berry, Craig Gidney, Mario Motta, Jarrod R McClean, and Ryan Babbush. Qubitization of arbitrary basis quantum chem- istry leveraging sparsity and low rank factor- ization. Quantum, 3:208, 2019

    2019

  10. [16]

    Hamil- tonian simulation by qubitization

    Guang Hao Low and Isaac L Chuang. Hamil- tonian simulation by qubitization. Quantum, 3:163, 2019

    2019

  11. [17]

    Quan- tum signal processing

    Yonina C Eldar and Alan V Oppenheim. Quan- tum signal processing. IEEE Signal Processing Magazine, 19(6):12–32, 2002

    2002

  12. [18]

    Quantum singular value transformation and beyond: exponential im- provements for quantum matrix arithmetics

    Andr´ as Gily´ en, Yuan Su, Guang Hao Low, and Nathan Wiebe. Quantum singular value transformation and beyond: exponential im- provements for quantum matrix arithmetics. In Proceedings of the 51st annual ACM SIGACT symposium on theory of computing, pages 193– 204, 2019

    2019

  13. [19]

    The theory of variational hybrid quantum-classical algo- rithms

    Jarrod R McClean, Jonathan Romero, Ryan Babbush, and Al´ an Aspuru-Guzik. The theory of variational hybrid quantum-classical algo- rithms. New Journal of Physics, 18(2):023023, 2016. 21 Circuit T ruth T able Circuit T ruth T able q0 q1 q2 q0 q1 q2 q′ 0 q′ 1 q′ 2 0 0 0 0 0 0 0 ...

    2016

  14. [20]

    easy” to implement, i.e., simple unitaries with low- depth circuits, can be added to the Sigma ba- sis. This flexibility makes our approach truly a linear combination on “things

    with the addition of certain permutation matrices and was used to efficiently represent Carleman linearized Burgers’ equation. In gen- eral, we note that any matrix that is “easy” to implement, i.e., simple unitaries with low- depth circuits, can be added to the Sigma ba- sis....

  15. [21]

    Application of fermionic marginal constraints to hybrid quantum algorithms

    Nicholas C Rubin, Ryan Babbush, and Jarrod McClean. Application of fermionic marginal constraints to hybrid quantum algorithms. New Journal of Physics, 20(5):053020, 2018

    2018

  16. [23]

    Precise mea- surement of quantum observables with neural- network estimators

    Giacomo Torlai, Guglielmo Mazzola, Giuseppe Carleo, and Antonio Mezzacapo. Precise mea- surement of quantum observables with neural- network estimators. Physical Review Research, 2(2):022060, 2020

    2020

  17. [24]

    Variational quantum algorithm 22 for the poisson equation

    Hai-Ling Liu, Yu-Sen Wu, Lin-Chun Wan, Shi- Jie Pan, Su-Juan Qin, Fei Gao, and Qiao- Yan Wen. Variational quantum algorithm 22 for the poisson equation. Physical Review A, 104(2):022418, 2021

    2021

  18. [25]

    Quan- tum computation and quantum information

    Michael A Nielsen and Isaac L Chuang. Quan- tum computation and quantum information . Cambridge university press, 2010

    2010

  19. [26]

    Tensorized pauli decomposi- tion algorithm

    Lukas Hantzko, Lennart Binkowski, and Sab- hyata Gupta. Tensorized pauli decomposi- tion algorithm. Physica Scripta, 99(8):085128, 2024

    2024

  20. [27]

    Efficient quan- tum algorithm for dissipative nonlinear differ- ential equations

    Jin-Peng Liu, Herman Øie Kolden, Hari K Krovi, Nuno F Loureiro, Konstantina Triv- isa, and Andrew M Childs. Efficient quan- tum algorithm for dissipative nonlinear differ- ential equations. Proceedings of the National Academy of Sciences , 118(35):e2026805118, 2021

    2021

  21. [28]

    Po- tential quantum advantage for simulation of fluid dynamics

    Xiangyu Li, Xiaolong Yin, Nathan Wiebe, Jae- hun Chun, Gregory K Schenter, Margaret S Cheung, and Johannes M¨ ulmenst¨ adt. Po- tential quantum advantage for simulation of fluid dynamics. Physical Review Research, 7(1):013036, 2025

    2025

  22. [30]

    An efficient quantum algorithm for simulating polynomial dynamical systems

    Amit Surana, Abeynaya Gnanasekaran, and Tuhin Sahai. An efficient quantum algorithm for simulating polynomial dynamical systems. Quantum Information Processing, 23(3):105, 2024

    2024

  23. [31]

    Variational quantum framework for nonlinear pde constrained optimization us- ing carleman linearization

    Abeynaya Gnanasekaran, Amit Surana, and Hongyu Zhu. Variational quantum framework for nonlinear pde constrained optimization us- ing carleman linearization. Quantum Informa- tion & Computation, 25(3):260–289, 2025

    2025

  24. [32]

    An efficient decomposition of the carleman linearized burgers’ equation

    Reuben Demirdjian, Thomas Hogancamp, and Daniel Gunlycke. An efficient decomposition of the carleman linearized burgers’ equation. arXiv preprint arXiv:2505.00285, 2025

    2025

  25. [33]

    M. Fang, S. Fenner, F. Green, S. Homer, and Y. Zhang. Quantum lower bounds for fanout. Quantum Info. Comput., 6(1):46–57, jan 2006

    2006

  26. [34]

    Quantum circuit for multi-qubit toffoli gate with optimal resource

    Junhong Nie, Wei Zi, and Xiaoming Sun. Quantum circuit for multi-qubit toffoli gate with optimal resource. arXiv preprint arXiv:2402.05053, 2024

    2024 arXiv

  27. [35]

    N-body interactions between trapped ion qubits via spin-dependent squeezing

    Or Katz, Marko Cetina, and Christopher Mon- roe. N-body interactions between trapped ion qubits via spin-dependent squeezing. Physical Review Letters, 129(6):063603, 2022

    2022

  28. [36]

    Lecture notes on quantum algorithms for scientific computation, 2022

    Lin Lin. Lecture notes on quantum algorithms for scientific computation, 2022

    2022

  29. [37]

    Fable: Fast approximate quantum circuits for block- encodings

    Daan Camps and Roel Van Beeumen. Fable: Fast approximate quantum circuits for block- encodings. In 2022 IEEE International Confer- ence on Quantum Computing and Engineering (QCE), pages 104–113. IEEE, 2022

    2022

  30. [38]

    Explicit quantum circuits for block encodings of certain sparse matrices

    Daan Camps, Lin Lin, Roel Van Beeumen, and Chao Yang. Explicit quantum circuits for block encodings of certain sparse matrices. SIAM Journal on Matrix Analysis and Applications, 45(1):801–827, 2024

    2024

  31. [39]

    Block-encoding structured ma- trices for data input in quantum computing

    Christoph S¨ underhauf, Earl Campbell, and Joan Camps. Block-encoding structured ma- trices for data input in quantum computing. Quantum, 8:1226, 2024

    2024

  32. [40]

    Quantum signal processing and singular value transformation: Lecture 1, 2023

    Ewin Tang. Quantum signal processing and singular value transformation: Lecture 1, 2023

    2023

  33. [41]

    Circuit com- plexity of quantum access models for encod- ing classical data

    Xiao-Ming Zhang and Xiao Yuan. Circuit com- plexity of quantum access models for encod- ing classical data. npj Quantum Information, 10(1):42, 2024

    2024

  34. [42]

    On unitary dilations of contrac- tions

    JJ Sch¨ affer. On unitary dilations of contrac- tions. In Proc. Amer. Math. Soc, volume 6, page 322, 1955

    1955

  35. [43]

    Variational quantum framework for partial differential equation constrained optimization

    Amit Surana and Abeynaya Gnanasekaran. Variational quantum framework for partial differential equation constrained optimization. arXiv preprint arXiv:2405.16651, 2024

    2024 arXiv

Pith tools

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