Pith. sign in

REVIEW 3 major objections 6 minor 41 references

Comparing performance of variational quantum algorithm simulations on HPC systems

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 shared parser ports the same quantum optimization problem across simulators, and the resulting runs are limited by runtime rather than memory.

desk verdict Useful parser tool and benchmark data, but the 'same circuits' claim outruns the evidence because OpenQASM 2.0 can't represent the parameterized ansätze. read the letter →

arxiv 2507.17614 v1 pith:LJ3VK5KB submitted 2025-07-23 quant-ph cs.DC

Marco De Pascale , Tobias Valentin Bauer , Yaknan John Gambo , Mario Hernández Vera , Stefan Huber , Burak Mete , Amit Jamadagni , Amine Bentellis
show 3 more authors
Marita Oliv Luigi Iapichino Jeanette Miriam Lorenz
This is my paper · ORCID
classification quant-phcs.DC
keywords variationalquantumalgorithmscircuitsimulationHamiltonianintermediaterepresentationansatzparserHPCbenchmarkingVQEQAOAjobarrays
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

Variational quantum simulations are hard to compare because each software package builds Hamiltonians and ansätze differently, so seemingly identical runs differ from one simulator to the next. This paper argues that a generic intermediate representation—one list-based format for Hamiltonians and one circuit format for ansätze—can carry a single problem definition consistently into several state-vector simulators, and it demonstrates the porting on three use cases: the hydrogen molecule, MaxCut, and the traveling salesperson problem. Verification runs on different HPC systems show that the translated circuits agree on the discrete cost-function levels, with energy differences at or near numerical round-off, so the toolchain succeeds at its stated goal. The performance finding is that variational workloads of these sizes are limited by runtime, not by memory: a 20-qubit MaxCut state occupies only 16 MB, yet the per-trajectory time is long enough that CPU-thread scaling quickly saturates. The authors conclude that job arrays are the natural way to parallelize such workloads and that GPU-accelerated simulators offer a favorable balance of usability and performance.

What carries the argument

The central machinery is a pair of simulator-agnostic intermediate representations plus per-simulator parsers: a Hamiltonian IR storing qubit count, coefficients, and Pauli operators in binary symplectic vectors, and an ansatz IR using OpenQASM v2.0 circuits. These ensure that every simulator receives the same circuit and the same operator, so differences in gate sets, native circuit construction, or nuclear-repulsion constants cannot masquerade as algorithmic differences. The performance analysis then leans on a "quantum time" metric—the median wall time spent actually executing the parameterized circuits, with optimizer time treated as negligible because the optimization takes a minor fraction of the total execution time.

What would settle it

Profile a larger MaxCut instance (28–30 qubits) or a deeper QAOA circuit with separate timers for circuit execution and classical optimizer calls; if optimizer time exceeds a few percent of wall time, the paper's quantum-time metric and its scaling conclusions fail to hold.

Watch

Extended reading notes

Core claim

The paper's central claim has two parts. First, a purpose-built parser that represents a Hamiltonian as a flat list of qubit count, coefficients, and Pauli words encoded via binary symplectic vectors, and an ansatz as an OpenQASM v2.0 circuit string, lets the same quantum problem be translated faithfully among the compared simulators. Second, measured trajectory quantum times for 15- and 20-qubit MaxCut instances show that the workload is compute-time-bound: on a full CPU node, quantum times range from about 0.07 to 0.93 seconds per trajectory for the 20-qubit instance while the state vector occupies only 16 MB, and speedup curves flatten or degrade beyond roughly 24 to 48 threads. The paper reports that the optimization component is a minor fraction of total execution time, that GPU-accelerated runs show the shortest quantum times with scaling factors far below the worst-case 32-fold growth of the state space, and that the cost-function values from all simulators align on discrete levels, which validates both the parser and the consistency of the results.

Load-bearing premise

The load-bearing premise is that the classical optimizer's run time stays a minor fraction of total execution time, so the reported "quantum time" truly measures simulator performance.

Editorial extensions

If this is right

  • A user can define a variational quantum problem once and run the identical circuit and Hamiltonian on multiple simulators, making cross-simulator benchmarking and result verification a mechanical step rather than a reimplementation effort.
  • Since a 20-qubit MaxCut state fits in 16 MB while single-trajectory runtimes are seconds, further scaling within a node will hit time-to-solution limits long before memory limits, so speedups from additional CPU threads will remain sublinear.
  • Bundling many random-initialization trajectories into job arrays is a practical way to use idle cores, partially compensating for the weak parallelism inside each individual circuit.
  • GPU-accelerated backends show the shortest quantum times and scale well below the worst-case growth of the state space, suggesting that variational simulation throughput is more favorable on GPU-accelerated machines than on CPU-only nodes.
  • Cross-simulator agreement of cost-function levels at round-off precision makes the parsed problem definitions a reliable baseline for checking new simulators or future hardware backends.

Reading between the lines

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

  • If classical optimizer cost grows with problem size or landscape difficulty, the assumption that optimization time is negligible would need revisiting; a simple extension is to report both circuit-only and end-to-end times when benchmarking larger instances.
  • The same Hamiltonian and ansatz intermediate representations could be extended beyond state-vector simulators to tensor-network or hardware backends, preserving the "identical circuits" guarantee while changing the performance profile.
  • Because the runtime bottleneck already dominates at only 20 qubits, pushing to 28–30 qubits on one node should reveal whether distributed-memory state-vector simulation becomes necessary before memory itself becomes the constraint.
  • OpenQASM v3 parametric circuits would let each trajectory vary parameters without re-parsing full circuits, likely reducing per-trajectory overhead and shifting the measured speedups.
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 manuscript introduces a parser tool that maps Hamiltonian and ansatz definitions into simulator-agnostic intermediate representations (a Pauli-word list for Hamiltonians and OpenQASM 2.0 for ansätze) to port variational quantum algorithm workloads across seven simulators on three HPC systems. Three use cases are benchmarked: H2 VQE on SuperMUC-NG, MaxCut QAOA on CPU and GPU systems, and TSP VQE as a consistency check. The authors report vectorization and memory-stall profiles, container versus bare-metal timing, CPU and GPU scaling behavior, and discrete cost-function levels across simulators. They conclude that the parser successfully ensures that the same circuits run on all simulators and that variational algorithms are limited by long runtimes relative to their small memory footprints.

Significance. If the parser equivalence were rigorously established, this would be a useful community tool for cross-simulator benchmarking, and the performance observations would inform how VQA workloads are scheduled on HPC resources. The paper has concrete strengths: it uses a range of real HPC systems, collects 1000 random initial parameter sets per configuration, includes a container versus bare-metal comparison, and performs a cross-simulator consistency check. However, the central claims currently depend on two unverified assumptions: that the OpenQASM 2.0-based parser passes identical parameterized circuits to all simulators, and that optimizer time is negligible. Both need additional evidence before the quantitative comparisons can be taken as definitive.

major comments (3)
  1. [Section III and Section VII-C] The claim that the parser ensures 'the same circuits are running across different simulators' is not established for parameterized ansätze. The ansatz IR is OpenQASM 2.0, and the paper itself states that parametric-circuit handling is future work. Since VQE and QAOA require parameterized circuits, the tool must concretely bind angles before each evaluation, but the manuscript does not specify how this binding is performed or verify that gate sequences, qubit orderings, and parameter conventions are identical across simulators. The consistency check in Section VII-C compares only discrete cost-function levels from 1000 random restarts, which is a weak certificate because many inequivalent circuits, such as those with reordered commuting Pauli terms or different but equivalent gate decompositions, can yield the same energy-level structure. The TSP paragraph reports identical state vectors for CUDA-Q, but no systematic pairwise statevector or gate-level comparison across all simulators is presented. This gap is load-bearing for the central claim in Section VIII and for the fairness of the quantum-time comparisons.
  2. [Section VII and Table III] The central performance metric 'quantum time' is defined as the time to run the quantum circuit, with the assumption that 'the optimization time takes a minor fraction of the whole execution time and thus can be largely ignored.' No profiling data are presented that separate optimizer time from circuit execution time; Figure 1 reports whole-VQE elapsed time, and the H2 section uses single-circuit elapsed time as a proxy only after asserting that the optimization time is negligible. If optimizer overhead is not negligible, the reported quantum times and speedups would not isolate simulator performance. The authors should provide a time breakdown from profiling or otherwise justify the assumption for each simulator and problem size.
  3. [Table III, Figures 2-6, and Section VII-B5] The scaling conclusions are drawn from only two problem sizes (15 and 20 qubits), and Table III gives median quantum times without error bars or confidence intervals. The scaling factors Ssim in Section VII-B5 are ratios of medians with no uncertainty propagation. The claim in Section VIII that variational algorithms are limited by long runtimes relative to memory footprint would be stronger with additional problem sizes and with statistical measures of dispersion, particularly since Figure 6 shows substantial scatter for some simulators.
minor comments (6)
  1. [Equation (5)] The symbols sP and sD are not defined in the text; the authors should either define them as scaled penalty and distance factors or replace them with P and D to match Equation (4).
  2. [Section VII-B5] There is a typo in the text: 'cuQuanutm' should be 'cuQuantum'.
  3. [Section III] The description of the Hamiltonian IR would benefit from a concrete example showing how a Pauli word is encoded in binary symplectic vectors and how the parser maps this IR to each simulator's native format.
  4. [Figure 1 and Section VII-A1] Please clarify whether the elapsed time in Figure 1 includes the classical optimization loop; as written, it appears to conflict with the quantum-time definition in Section VII.
  5. [Section VII-C] The statement that CUDA-Q does not converge to the same values but that 'the actual state vectors for the optimization results are identical' needs one more sentence explaining how identical state vectors can produce a 10^-5 energy difference, since this bears on the consistency of expectation-value computations.
  6. [Section III] The manuscript states the parser software will soon be made available via GitLab and as a standalone Python package; for reproducibility of the benchmark numbers, a public repository or an appendix with the core mapping logic would be helpful at submission time.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the parser equivalence claim is checked against independent simulator backends; the self-citations are not load-bearing.

full rationale

The paper's central claim is that its Hamiltonian and ansatz parsers port the same problem definition to several independent simulators. The Hamiltonian IR is built from the PennyLane quantum chemistry database, and the ansatz IR is OpenQASM 2.0; neither is defined in terms of the paper's own output metrics. The consistency check in Section VII-C compares cost-function levels and state vectors across different simulator packages, and for the TSP case reports that CUDA-Q's state vectors are identical to the others, providing independent cross-backend evidence rather than a fitted prediction. The performance claims are direct elapsed-time measurements; the 'quantum time' approximation that optimizer cost is negligible is a stated modeling assumption, not a quantity fitted to reproduce the reported speedups. Self-citations [16] and [17] motivate container deployment tests and benchmarking context, but the parser's correctness does not rest on them; reference [29] supplies the external code basis for the parser. The validation is mutual consistency rather than absolute physical correctness, and the OpenQASM 2.0 parametric-circuit handling is a genuine limitation, but these are correctness risks, not circular reductions. No step in the derivation chain is equivalent by construction to its own input.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the parser design and the performance methodology. No fitted parameters drive the main conclusions, but several choices such as the TSP penalty factor and QAOA layer count are not fully specified. The main unverified premise is the negligible optimizer overhead, which underpins the core performance metric.

free parameters (3)
  • TSP penalty factor P
    In Eqs. (4) and (5), P is a penalty weight for the TSP constraints. Its value is not reported, and it affects the Hamiltonian and thus the optimization landscape and runtime. A reader cannot reproduce the TSP experiments without it.
  • Number of QAOA layers j
    Section II describes QAOA with a generic layer count j, but the number used in the MaxCut simulations is not reported. Circuit depth directly affects runtime, making the performance numbers hard to reproduce or compare.
  • TSP instance size = 4 nodes
    The text states '4 node instances, i.e. 9 qubits' but does not give the distance matrix or the exact graph. The problem instance is a design choice that influences runtime and solution structure.
assumptions (4)
  • domain assumption Born-Oppenheimer approximation and STO-3G basis for the H2 Hamiltonian
    Section II-B1 uses this standard quantum chemistry setup to define the H2 use case. The parser then maps this Hamiltonian, so the validity of the benchmark depends on this assumption being acceptable for the problem.
  • domain assumption Jordan-Wigner transformation and UCCSD ansatz are valid for H2 VQE
    Section II-B1 chooses these mappings and the UCCSD circuit. The performance results for the H2 use case depend on this specific algorithm choice.
  • domain assumption State vector simulation is representative of VQA performance on HPC
    The study is scoped to state vector simulators (Section VI). The general conclusions about scaling are only about this simulation method, and may not transfer to tensor network or other approximate methods.
  • ad hoc to paper Optimizer time is negligible relative to quantum circuit execution
    Section VII states this without showing profiling data. The 'quantum time' metric and all speedup plots rely on this assumption. If it fails, the performance comparisons could be biased.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Comparing performance of variational quantum algorithm simulations on HPC systems." pith.science (2026). https://pith.science/paper/LJ3VK5KB

@misc{pith2026250717614,
  author       = {Pith},
  title        = {Pith review of: Comparing performance of variational quantum algorithm simulations on HPC systems},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LJ3VK5KB}},
  note         = {Machine review of arXiv:2507.17614}
}
read the original abstract

Variational quantum algorithms are of special importance in the research on quantum computing applications because of their applicability to current Noisy Intermediate-Scale Quantum (NISQ) devices. The main building blocks of these algorithms (among them, the definition of the Hamiltonian and of the ansatz, the optimizer) define a relatively large parameter space, making the comparison of results and performance between different approaches and software simulators cumbersome and prone to errors. In this paper, we employ a generic description of the problem, in terms of both Hamiltonian and ansatz, to port a problem definition consistently among different simulators. Three use cases of relevance for current quantum hardware (ground state calculation for the Hydrogen molecule, MaxCut, Travelling Salesman Problem) have been run on a set of HPC systems and software simulators to study the dependence of performance on the runtime environment, the scalability of the simulation codes and the mutual agreement of the physical results, respectively. The results show that our toolchain can successfully translate a problem definition between different simulators. On the other hand, variational algorithms are limited in their scaling by the long runtimes with respect to their memory footprint, so they expose limited parallelism to computation. This shortcoming is partially mitigated by using techniques like job arrays. The potential of the parser tool for exploring HPC performance and comparisons of results of variational algorithm simulations is highlighted.

Figures

Figures reproduced from arXiv: 2507.17614 by the authors.

Figure 1
Figure 1. Bar plot of the elapsed time to execute the VQE algorithms of [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 3
Figure 3. Speedup plot for MaxCut simulations using Qiskit on SNG in both the [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Speedup of simulations for the MaxCut problem using the Intel-QS [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figures from the paper (3 more)
Figure 5
Figure 5. Figure 5: Speedup of simulations for the MaxCut problem using the Intel-QS on [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Box plots of the q time for different simulators and problem sizes on the BDAI system. We removed the whiskers from the Qiskit and CUDA-Q boxplots to improve the appearance. the underlying simulation backend cuQuanutm introduce some overhead. A potential analysis of th…
Figure 8
Figure 8. Figure 8: Centers of the levels of the cost function values for the different [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 22 canonical work pages

  1. [1]

    Near-Term Quantum Computing Techniques: Variational Quantum Algorithms, Error Mitigation, Circuit Compilation, Benchmarking and Classical Simulation

    H.-L. Huang, X.-Y . Xu, C. Guo, G. Tian, S.-J. Wei, X. Sun, W.-S. Bao, and G.-L. Long, “Near-Term Quantum Computing Techniques: Variational Quantum Algorithms, Error Mitigation, Circuit Compilation, Benchmarking and Classical Simulation,” arXiv e-prints, p. arXiv:2211.08737, 2022. [Online]. Available: https://ui.adsabs. harvard.edu/abs/2022arXiv221108737H

    work page Pith review arXiv 2022

  2. [2]

    A Herculean task: Classical simulation of quantum computers,

    X. Xu, S. Benjamin, J. Sun, X. Yuan, and P. Zhang, “A Herculean task: Classical simulation of quantum computers,” arXiv e-prints, p. arXiv:2302.08880, 2023. [Online]. Available: https: //ui.adsabs.harvard.edu/abs/2023arXiv230208880X

    arXiv 2023

  3. [3]

    Simulating Quantum Computations on Classical Machines: A Survey,

    K. Young, M. Scese, and A. Ebnenasir, “Simulating Quantum Computations on Classical Machines: A Survey,” arXiv e-prints, p. arXiv:2311.16505, 2023. [Online]. Available: https://ui.adsabs.harvard. edu/abs/2023arXiv231116505Y

    arXiv 2023

  4. [4]

    Massively parallel quantum computer simulator, eleven years later,

    H. De Raedt, F. Jin, D. Willsch, M. Willsch, N. Yoshioka, N. Ito, S. Yuan, and K. Michielsen, “Massively parallel quantum computer simulator, eleven years later,” Computer Physics Communications, vol. 237, pp. 47–61, 2019. [Online]. Available: https://ui.adsabs.harvard. edu/abs/2019CoPhC.237...47D

    work page 2019

  5. [5]

    Quest and high performance simulation of quantum computers,

    T. Jones, A. Brown, I. Bush, and S. C. Benjamin, “Quest and high performance simulation of quantum computers,” Scientific Reports, vol. 9, no. 1, Jul. 2019. [Online]. Available: http: //dx.doi.org/10.1038/s41598-019-47174-9

    doi:10.1038/s41598-019-47174-9 2019

  6. [6]

    Intel quantum simulator: a cloud-ready high-performance simulator of quantum circuits,

    G. G. Guerreschi, J. Hogaboam, F. Baruffa, and N. P. D. Sawaya, “Intel quantum simulator: a cloud-ready high-performance simulator of quantum circuits,” Quantum Science and Technology, vol. 5, no. 3, p. 034007, 2020. [Online]. Available: http://dx.doi.org/10.1088/ 2058-9565/ab8505

    work page 2020

  7. [7]

    Variational quantum algorithms,

    M. Cerezo, A. Arrasmith, R. Babbush, S. C. Benjamin, S. Endo, K. Fujii, J. R. McClean, K. Mitarai, X. Yuan, L. Cincio, and P. J. Coles, “Variational quantum algorithms,” Nature Reviews Physics, vol. 3, no. 9, pp. 625–644, 2021. [Online]. Available: https://doi.org/10.1038/s42254-021-00348-9

    doi:10.1038/s42254-021-00348-9 2021

  8. [8]

    Vqe method: a short survey and recent developments,

    D. A. Fedorov, B. Peng, N. Govind, and Y . Alexeev, “Vqe method: a short survey and recent developments,” Materials Theory, vol. 6, no. 1, Jan. 2022. [Online]. Available: http://dx.doi.org/10.1186/ s41313-021-00032-6

    work page 2022

Show all 41 references
  1. [9]

    Quantum computers for high-performance computing,

    T. S. Humble, A. McCaskey, D. I. Lyakh, M. Gowrishankar, A. Frisch, and T. Monz, “Quantum computers for high-performance computing,” IEEE Micro, vol. 41, no. 5, pp. 15–23, 2021

    2021

  2. [10]

    Accel- erating hpc with quantum computing: It is a software challenge too,

    M. Schulz, M. Ruefenacht, D. Kranzlmuller, and L. Schulz, “Accel- erating hpc with quantum computing: It is a software challenge too,” Computing in Science & Engineering, vol. 24, no. 04, pp. 60–64, 2022

    2022

  3. [11]

    Integration of Quantum Accelerators with High Performance Computing – A Review of Quantum Programming Tools,

    A. Elsharkawy, X.-T. M. To, P. Seitz, Y . Chen, Y . Stade, M. Geiger, Q. Huang, X. Guo, M. Arslan Ansari, C. B. Mendl, D. Kranzlmüller, and M. Schulz, “Integration of Quantum Accelerators with High Performance Computing – A Review of Quantum Programming Tools,” arXiv e-prints,...

    2023

  4. [12]

    Toward a Unified Hybrid HPCQC Toolchain,

    P. Seitz, A. Elsharkawy, X.-T. M. To, and M. Schulz, “Toward a Unified Hybrid HPCQC Toolchain,” arXiv e-prints, p. arXiv:2309.01661, 2023. [Online]. Available: https://ui.adsabs.harvard. edu/abs/2023arXiv230901661S

    2023 arXiv

  5. [13]

    Towards the munich quantum software stack: Enabling efficient access and tool support for quantum computers,

    M. Schulz, L. Schulz, M. Ruefenacht, and R. Wille, “Towards the munich quantum software stack: Enabling efficient access and tool support for quantum computers,” in 2023 IEEE International Conference on Quantum Computing and Engineering (QCE). IEEE, Sep. 2023. [Online]. Availa...

    2023

  6. [14]

    A stabilizer framework for Contextual Subspace VQE and the noncontextual projection ansatz,

    T. J. Weaving, A. Ralli, W. M. Kirby, A. Tranter, P. J. Love, and P. V . Coveney, “A stabilizer framework for Contextual Subspace VQE and the noncontextual projection ansatz,” arXiv e-prints, p. arXiv:2204.02150, 2022. [Online]. Available: https://ui.adsabs.harvard. edu/abs/20...

    2022 arXiv

  7. [15]

    Evaluating the impact of noise on the performance of the Variational Quantum Eigensolver,

    M. Oliv, A. Matic, T. Messerer, and J. M. Lorenz, “Evaluating the impact of noise on the performance of the Variational Quantum Eigensolver,” arXiv e-prints, p. arXiv:2209.12803, 2022. [Online]. Available: https://ui.adsabs.harvard.edu/abs/2022arXiv220912803O

    2022 arXiv

  8. [16]

    Deploying Containerized QuantEx Quantum Simu- lation Software on HPC Systems,

    D. Brayford, J. Brennan, M. Allalen, K. Hanley, L. Iapichino, L. ORior- dan, and N. Moran, “Deploying Containerized QuantEx Quantum Simu- lation Software on HPC Systems,” arXiv e-prints, p. arXiv:2110.05162, Oct. 2021

    2021 arXiv

  9. [17]

    Benchmarking quantum computer simulation software packages,

    A. Jamadagni, A. M. Läuchli, and C. Hempel, “Benchmarking quantum computer simulation software packages,” arXiv e-prints, p. arXiv:2401.09076, 2024. [Online]. Available: https://ui.adsabs.harvard. edu/abs/2024arXiv240109076J

    2024 arXiv

  10. [18]

    qHiPSTER: The Quantum High Performance Software Testing Environment,

    M. Smelyanskiy, N. P. D. Sawaya, and A. Aspuru-Guzik, “qHiPSTER: The Quantum High Performance Software Testing Environment,” arXiv e-prints, 2016. [Online]. Available: https://ui.adsabs.harvard.edu/\#abs/ 2016arXiv160107195S

    2016

  11. [19]

    Hybrid quantum programming with PennyLane Lightning on HPC platforms,

    A. Asadi, A. Dusko, C.-Y . Park, V . Michaud-Rioux, I. Schoch, S. Shu, T. Vincent, and L. J. O’Riordan, “Hybrid quantum programming with PennyLane Lightning on HPC platforms,” arXiv e-prints, p. arXiv:2403.02512, 2024. [Online]. Available: https: //ui.adsabs.harvard.edu/abs/20...

    2024 arXiv

  12. [20]

    A variational eigenvalue solver on a photonic quantum processor,

    A. Peruzzo, J. McClean, P. Shadbolt, M.-H. Yung, X.-Q. Zhou, P. J. Love, A. Aspuru-Guzik, and J. L. O’brien, “A variational eigenvalue solver on a photonic quantum processor,”Nature communications, vol. 5, no. 1, p. 4213, 2014

    2014

  13. [21]

    A Quantum Approximate Optimization Algorithm,

    E. Farhi, J. Goldstone, and S. Gutmann, “A Quantum Approximate Optimization Algorithm,” Nov. 2014

    2014

  14. [22]

    From the Quantum Approximate Optimization Algorithm to a Quantum Alternating Operator Ansatz,

    S. Hadfield, Z. Wang, B. O’Gorman, E. G. Rieffel, D. Venturelli, and R. Biswas, “From the Quantum Approximate Optimization Algorithm to a Quantum Alternating Operator Ansatz,” Algorithms, vol. 12, no. 2, p. 34, Feb. 2019

    2019

  15. [23]

    Nocedal and S

    J. Nocedal and S. J. Wright, Numerical Optimization. Springer New York, NY , 2006

    2006

  16. [24]

    Alternative coupled- cluster ansätze ii. the unitary coupled-cluster method,

    R. J. Bartlett, S. A. Kucharski, and J. Noga, “Alternative coupled- cluster ansätze ii. the unitary coupled-cluster method,” Chemical Physics Letters, vol. 155, no. 1, pp. 133–140, 1989. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0009261489873725

    1989

  17. [25]

    Jordan and E

    P. Jordan and E. P. Wigner, Über das paulische äquivalenzverbot. Springer, 1993

    1993

  18. [26]

    M. A. Nielsen and I. L. Chuang, Quantum computation and quantum information. Cambridge University Press, 2010

    2010

  19. [27]

    Differentiable quantum computational chemistry with pennylane,

    J. M. Arrazola, S. Jahangiri, A. Delgado, J. Ceroni, J. Izaac, A. Száva, U. Azad, R. A. Lang, Z. Niu, O. Di Matteo et al., “Differentiable quantum computational chemistry with pennylane,” arXiv preprint arXiv:2111.09967, 2021

    2021 arXiv

  20. [28]

    Openqasm 3: A broader and deeper quantum assembly language,

    A. Cross, A. Javadi-Abhari, T. Alexander, N. De Beaudrap, L. S. Bishop, S. Heidel, C. A. Ryan, P. Sivarajah, J. Smolin, J. M. Gambetta et al., “Openqasm 3: A broader and deeper quantum assembly language,”ACM Transactions on Quantum Computing, vol. 3, no. 3, pp. 1–50, 2022

    2022

  21. [29]

    Benchmarking quantum computer simulation software packages: State vector simulators,

    A. J. Gangapuram, A. M. Läuchli, and C. Hempel, “Benchmarking quantum computer simulation software packages: State vector simulators,” SciPost Phys. Core, vol. 7, p. 075, 2024. [Online]. Available: https://scipost.org/10.21468/SciPostPhysCore.7.4.075

    2024 doi

  22. [30]

    Nvidia cuquantum sdk,

    The cuQuantum development team, “Nvidia cuquantum sdk,” nov 2023

    2023

  23. [31]

    Qiskit: An open-source framework for quantum computing,

    Qiskit contributors, “Qiskit: An open-source framework for quantum computing,” 2023

    2023

  24. [32]

    Pennylane: Automatic differentiation of hybrid quantum-classical com- putations,

    V . Bergholm, J. Izaac, M. Schuld, C. Gogolin, S. Ahmed, V . Ajith, M. S. Alam, G. Alonso-Linaje, B. AkashNarayanan, A. Asadi et al., “Pennylane: Automatic differentiation of hybrid quantum-classical com- putations,” arXiv preprint arXiv:1811.04968, 2018

    2018 arXiv

  25. [33]

    Developers, “Cirq,” Jul

    C. Developers, “Cirq,” Jul. 2023. [Online]. Available: https://doi.org/ 10.5281/zenodo.8161252

    2023 doi

  26. [34]

    myqlm - quantum python package,

    The myQLM development team, “myqlm - quantum python package,” 2024, accessed 05.10.2024. [Online]. Available: https://myqlm.github.io

    2024

  27. [35]

    Cuda quantum

    The CUDA Quantum development team, “Cuda quantum.” [Online]. Available: https://github.com/NVIDIA/cuda-quantum

  28. [36]

    Charliecloud: unprivileged containers for user-defined software stacks in hpc,

    R. Priedhorsky and T. Randles, “Charliecloud: unprivileged containers for user-defined software stacks in hpc,” in Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, ser. SC ’17. New York, NY , USA: Association for Com...

    2017

  29. [37]

    Efficient calculation of gradients in classical simulations of variational quantum algorithms,

    T. Jones and J. Gacon, “Efficient calculation of gradients in classical simulations of variational quantum algorithms,” 2020. [Online]. Available: https://arxiv.org/abs/2009.02823

    2020 arXiv

  30. [38]

    Second order differentiation using adjoint differ- entiation

    “Second order differentiation using adjoint differ- entiation.” [Online]. Available: https://discuss.pennylane.ai/t/ second-order-differentiation-using-adjoint-differentiation/2705

  31. [39]

    Nvidia cuda software and gpu parallel computing architecture,

    D. Kirk, “Nvidia cuda software and gpu parallel computing architecture,” in Proceedings of the 6th International Symposium on Memory Management, ser. ISMM ’07. New York, NY , USA: Association for Computing Machinery, 2007, p. 103–104. [Online]. Available: https://doi.org/10.11...

    2007

  32. [40]

    The Santa Barbara Cluster Comparison Project: A Comparison of Cosmological Hydrodynamics Solutions,

    C. S. Frenk, S. D. M. White, P. Bode, J. R. Bond, G. L. Bryan, R. Cen, H. M. P. Couchman, A. E. Evrard, N. Gnedin, A. Jenkins, A. M. Khokhlov, A. Klypin, J. F. Navarro, M. L. Norman, J. P. Ostriker, J. M. Owen, F. R. Pearce, U.-L. Pen, M. Steinmetz, P. A. Thomas, J. V . Villum...

  33. [1999]

    Available: http://esoads.eso.org/cgi-bin/nph-bib_query? bibcode=1999ApJ...525..554F&db_key=AST

    [Online]. Available: http://esoads.eso.org/cgi-bin/nph-bib_query? bibcode=1999ApJ...525..554F&db_key=AST

Pith tools

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