pith. sign in

arxiv: 2606.31351 · v1 · pith:XTA2HNPSnew · submitted 2026-06-30 · 🪐 quant-ph

A Quantum-Classical Surrogate Model for the Collision Operator of the Lattice Boltzmann Method

Lukas C. Birk , David M. Wawrzyniak , Josef M. Winter , Steffen J. Schmidt , Thomas Indinger , Christian F. Jan{\ss}en , Nikolaus A. Adams This is my paper

Pith reviewed 2026-07-01 05:33 UTC · model grok-4.3

classification 🪐 quant-ph
keywords quantum machine learninglattice Boltzmann methodBGK collision operatorparameterized quantum circuitsdata re-uploadinghybrid quantum-classicalfluid dynamicssurrogate modeling
0
0 comments X

The pith

A single parameterized quantum circuit with data re-uploading approximates the full range of BGK collision operators without retraining.

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

The paper develops a hybrid quantum-classical surrogate to handle the nonlinear collision step in lattice Boltzmann simulations. A parameterized quantum circuit uses data re-uploading to represent the Bhatnagar-Gross-Krook operator across every admissible relaxation value after one training pass. This sidesteps the non-unitary operations that hinder fully quantum fluid solvers. The model is checked against the classical operator on the Taylor-Green vortex for dissipation and the double shear layer for instability growth. Results show the surrogate reproduces classical flow evolution at high accuracy while linking standard circuit metrics to task-specific performance.

Core claim

The surrogate recovers the complete Bhatnagar-Gross-Krook (BGK) collision dynamics across the full physically admissible range of relaxation without retraining, built on the ability of parameterized quantum circuits to implement partial Fourier series with data re-uploading to extend representable frequencies.

What carries the argument

Parameterized quantum circuit with data re-uploading that implements partial Fourier series to represent the collision operator for varying relaxation parameters.

If this is right

  • The hybrid model reproduces energy dissipation rates in the Taylor-Green vortex to high accuracy.
  • The surrogate captures shear-driven instabilities and nonlinear evolution in the double shear layer.
  • Expressibility, entanglement, and effective dimension of the circuit can be related directly to surrogate error on the collision task.
  • Specific architectural choices in the circuit determine the achieved approximation accuracy.
  • The approach offloads non-unitary collision operations while retaining classical handling of the streaming step.

Where Pith is reading between the lines

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

  • The same circuit structure might extend to other collision models such as multiple-relaxation-time variants if the Fourier representation generalizes.
  • Integration into larger-scale engineering simulations could test whether the surrogate reduces wall-clock time relative to classical collision evaluations.
  • The re-assessment of variational metrics suggests similar task-specific validation could be applied to quantum surrogates in other transport or kinetic problems.
  • Demonstration on three-dimensional or turbulent regimes would clarify whether the current accuracy holds beyond the two-dimensional benchmarks used.

Load-bearing premise

A single trained parameterized quantum circuit with data re-uploading can represent the BGK collision operator accurately for any relaxation parameter in the admissible range.

What would settle it

Apply the surrogate to a third benchmark flow such as lid-driven cavity flow at a relaxation parameter outside the training distribution and compare the resulting velocity field or energy spectrum against the classical BGK operator.

Figures

Figures reproduced from arXiv: 2606.31351 by Christian F. Jan{\ss}en, David M. Wawrzyniak, Josef M. Winter, Lukas C. Birk, Nikolaus A. Adams, Steffen J. Schmidt, Thomas Indinger.

Figure 1
Figure 1. Figure 1: Quantum circuit ansätze for an N-qubit circuit: (a) Line, (b) Parallel, (c) Super￾Parallel. The initial entangling layer W(θ (0)) is followed by L layers, each comprising encoding gates S(xn) and entangling layers W(θ (l) ). multi-dimensional data representation is realised via non-commuting rota￾tions about different axes of the Bloch sphere. Similarly, the Parallel Ansatz (PA) applies a distinct encoding… view at source ↗
Figure 2
Figure 2. Figure 2: Quantum circuits illustrating two entangling-layer architectures: (a) the Basic [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Gate counts (single-qubit, two-qubit, total), circuit depths, and gate density of [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: The KL divergence DKL is computed for various VQC variants (see fig. 3 for notation). Since the Haar reference fidelity distribution varies with N, DKL results are not directly comparable across N. Across all architectures, LSW is the primary driver of expressibility, while additional LW layers contribute only marginally beyond the first. Both PA BEL and PA SEL saturate with increasing depth, with the SEL … view at source ↗
Figure 5
Figure 5. Figure 5: Entangling tested for different entanglement variants with varying numbers of [PITH_FULL_IMAGE:figures/full_fig_p018_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Effective dimension (ED) and parameter-normalised ED for the PA and SPA [PITH_FULL_IMAGE:figures/full_fig_p019_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Box-and-whisker plots illustrating the relative error distributions for two inde [PITH_FULL_IMAGE:figures/full_fig_p021_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Relative error of SEL1 L6 as a function of [PITH_FULL_IMAGE:figures/full_fig_p022_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Relative error of SEL1 L6 as a function of training range [PITH_FULL_IMAGE:figures/full_fig_p023_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Extended circuit for BGK relaxation. The VQC outputs the equilibrium [PITH_FULL_IMAGE:figures/full_fig_p027_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Relative error of the trained SEL1 L6 surrogate for all 19 discrete velocity [PITH_FULL_IMAGE:figures/full_fig_p028_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Velocity field (ω = 1.25) of a three-dimensional TGV at Re ≈ 1, shown at time instances t = 0s, t = 0.25s, t = 0.5s, and t = 1.25s. DOC surrogate behaves comparably to the non-compressed variant, confirm￾ing that the single trained circuit is sufficiently accurate when applied to all directions within an orbit via this deterministic rotation scheme. Notably, orbit compression has a stabilising effect on t… view at source ↗
Figure 13
Figure 13. Figure 13: Non-dimensional kinetic energy over time for the three-dimensional Taylor [PITH_FULL_IMAGE:figures/full_fig_p031_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Relative error of the trained SEL1 L6 surrogate for all 9 discrete velocity [PITH_FULL_IMAGE:figures/full_fig_p031_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Velocity magnitude |u| (left column) and pointwise absolute error ei between surrogate and BGK reference (right column) for the Double Shear Layer at Re = 100 (top), Re = 400 (middle), and Re = 1000 (bottom), each shown at t = 10 s. lower peak compared to Re = 400 reflects the more spatially concentrated gradients, with a larger fraction of the domain remains in a moderate flow state. The quantitative agr… view at source ↗
Figure 16
Figure 16. Figure 16: Time evolution of non-dimensional kinetic energy [PITH_FULL_IMAGE:figures/full_fig_p034_16.png] view at source ↗
read the original abstract

We introduce a hybrid approach utilising a quantum machine learning surrogate model to approximate the non-linear collision dynamics of the LBM. It effectively offloads the non-unitary operations that challenge pure quantum solvers. The expressivity of the surrogate is built on the ability of parameterised quantum circuits to implement partial Fourier series, with data re-uploading extending the spectrum of representable frequencies. Unlike previous approaches with a fixed relaxation parameter, the surrogate recovers the complete Bhatnagar-Gross-Krook (BGK) collision dynamics across the full physically admissible range of relaxation without retraining. We reassess the relevance of standard variational quantum circuit (VQC) metrics, including expressibility, entanglement, and effective dimension, by relating them directly to task-specific surrogate performance and identifying the key architectural parameters that determine approximation accuracy. The proposed surrogate is validated against the classical BGK collision operator using established benchmark problems, including the Taylor-Green vortex for evaluating energy dissipation and the double shear layer for assessing shear-driven instabilities and nonlinear flow evolution. Our results demonstrate that the hybrid model achieves high accuracy and generalisability while closely replicating classical solutions. These findings suggest that hybrid quantum-classical strategies offer a practical path toward realising the potential of quantum computing in fluid engineering.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript introduces a hybrid quantum-classical surrogate for the Lattice Boltzmann Method collision operator, employing a parameterized quantum circuit (PQC) with data re-uploading to approximate the full Bhatnagar-Gross-Krook (BGK) operator f' = f - (f - f_eq)/tau. It claims that a single fixed trained model recovers the complete BGK dynamics for arbitrary relaxation parameters across the physically admissible range without retraining, reassesses standard VQC metrics (expressibility, entanglement, effective dimension) against task performance, and validates the approach on the Taylor-Green vortex (energy dissipation) and double shear layer (shear instabilities) benchmarks, asserting high accuracy and generalisability.

Significance. If the central generalization claim holds with quantitative support, the work would demonstrate a practical hybrid strategy for handling non-unitary collision steps in quantum LBM solvers and could advance quantum machine learning applications in computational fluid dynamics by linking VQC architectural choices directly to surrogate fidelity.

major comments (2)
  1. [Abstract] Abstract: the claim that the surrogate 'recovers the complete BGK collision dynamics across the full physically admissible range of relaxation without retraining' is load-bearing for the central contribution yet rests only on validation for the Taylor-Green vortex and double shear layer at specific (unspecified) tau values; no parameter sweeps, continuous tau variation tests, or out-of-distribution checks are described to substantiate the 'arbitrary' and 'complete' qualifiers.
  2. [Abstract] Abstract and validation description: the assertions of 'high accuracy and generalisability' are unsupported by any reported quantitative error metrics, training details, loss values, or error bars comparing the surrogate output to the classical BGK operator.
minor comments (1)
  1. [Abstract] Abstract: the phrase 'closely replicating classical solutions' is vague without reference to specific figures, tables, or error norms.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We address each major point below and will revise the manuscript to strengthen the presentation of our results.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim that the surrogate 'recovers the complete BGK collision dynamics across the full physically admissible range of relaxation without retraining' is load-bearing for the central contribution yet rests only on validation for the Taylor-Green vortex and double shear layer at specific (unspecified) tau values; no parameter sweeps, continuous tau variation tests, or out-of-distribution checks are described to substantiate the 'arbitrary' and 'complete' qualifiers.

    Authors: We agree that explicit parameter sweeps and out-of-distribution tests for tau are not described in the current validation sections. The surrogate is trained on data sampled across the admissible tau range to support generalization, but to fully substantiate the claim we will add a dedicated analysis with continuous tau sweeps and error metrics in the revised manuscript. revision: yes

  2. Referee: [Abstract] Abstract and validation description: the assertions of 'high accuracy and generalisability' are unsupported by any reported quantitative error metrics, training details, loss values, or error bars comparing the surrogate output to the classical BGK operator.

    Authors: Quantitative metrics, training details and error comparisons are included in the results section of the full manuscript. To address the concern directly in the abstract, we will revise it to report key quantitative error metrics, loss values and error bars. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper trains a parameterized quantum circuit surrogate on classical BGK data and validates performance by direct comparison to the classical collision operator on independent benchmark flows (Taylor-Green vortex, double shear layer). No quoted step defines a quantity in terms of itself, renames a fitted parameter as a prediction, or relies on a self-citation chain for the central claim. The generalization statement is an empirical assertion tested against external classical solutions rather than a reduction by construction. The derivation remains self-contained against the stated classical benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

Approach rests on standard quantum machine learning assumptions about circuit expressivity; no new entities postulated.

free parameters (1)
  • VQC parameters
    Circuit parameters are optimized to match classical BGK collision data for the surrogate task.
axioms (1)
  • domain assumption Parameterized quantum circuits with data re-uploading can implement partial Fourier series sufficient to represent BGK collision dynamics
    This is the stated basis for the surrogate expressivity in the abstract.

pith-pipeline@v0.9.1-grok · 5779 in / 1341 out tokens · 57874 ms · 2026-07-01T05:33:11.276987+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

63 extracted references · 54 canonical work pages · 5 internal anchors

  1. [1]

    P. W. Shor, Polynomial-time algorithms for prime factorization and dis- crete logarithms on a quantum computer, SIAM Journal on Computing 26 (5) (1997) 1484–1509. doi:10.1137/S0097539795293172. 36

    work page doi:10.1137/s0097539795293172 1997

  2. [2]

    L. K. Grover, A fast quantum mechanical algorithm for database search. URLhttp://arxiv.org/pdf/quant-ph/9605043

    work page internal anchor Pith review Pith/arXiv arXiv

  3. [3]

    A. W. Harrow, A. Hassidim, S. Lloyd, Quantum algorithm for linear systems of equations, Physical review letters 103 (15) (2009) 150502. doi:10.1103/PhysRevLett.103.150502. URLhttp://arxiv.org/pdf/0811.3171

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1103/physrevlett.103.150502 2009

  4. [4]

    N. Ray, T. Banerjee, B. Nadiga, S. Karra, Towards solving the navier- stokes equation on quantum computers, LA-UR-19-. URLhttp://arxiv.org/pdf/1904.09033

    work page internal anchor Pith review Pith/arXiv arXiv 1904

  5. [5]

    Steijl, Quantum algorithms for fluid simulations, in: F

    R. Steijl, Quantum algorithms for fluid simulations, in: F. Bulnes, V. N. Stavrou, O. Morozov, A. V. Bourdine (Eds.), Advances in Quantum Communication and Information, IntechOpen, 2020. doi:10.5772/intechopen.86685

    work page doi:10.5772/intechopen.86685 2020

  6. [6]

    Steijl, G

    R. Steijl, G. N. Barakos, Parallel evaluation of quantum algorithms for computational fluid dynamics, Computers & Fluids 173 (2018) 22–28. doi:10.1016/j.compfluid.2018.03.080

    work page doi:10.1016/j.compfluid.2018.03.080 2018

  7. [7]

    Gaitan, Finding flows of a navier–stokes fluid through quantum com- puting, npj Quantum Information 6 (1) (2020)

    F. Gaitan, Finding flows of a navier–stokes fluid through quantum com- puting, npj Quantum Information 6 (1) (2020). doi:10.1038/s41534-020- 00291-0

    work page doi:10.1038/s41534-020- 2020

  8. [8]

    F. Oz, R. K. S. S. Vuppala, K. Kara, F. Gaitan, Solving burgers’ equa- tion with quantum computing, Quantum Information Processing 21 (1) (2022). doi:10.1007/s11128-021-03391-8

    work page doi:10.1007/s11128-021-03391-8 2022

  9. [9]

    Yepez, Quantum lattice-gas model for computational fluid dynamics, Physical review

    J. Yepez, Quantum lattice-gas model for computational fluid dynamics, Physical review. E, Statistical, nonlinear, and soft matter physics 63 (4 Pt 2) (2001) 046702. doi:10.1103/PhysRevE.63.046702

    work page doi:10.1103/physreve.63.046702 2001

  10. [10]

    G. P. Berman, A. A. Ezhov, D. I. Kamenev, J. Yepez, Simulation of the diffusion equation on a type-ii quantum computer, Physical Review A 66 (1) (2002). doi:10.1103/PhysRevA.66.012310

    work page doi:10.1103/physreva.66.012310 2002

  11. [11]

    B. N. Todorova, R. Steijl, Quantum algorithm for the collisionless boltz- mann equation, Journal of Computational Physics 409 (2020) 109347. doi:10.1016/j.jcp.2020.109347. 37

    work page doi:10.1016/j.jcp.2020.109347 2020

  12. [12]

    Alouges, F

    F. Alouges, F. de Vuyst, G. Le Coq, E. Lorin, The reservoir technique: a way to make godunov-type schemes zero or very low diffuse. application to colella–glaz solver, European Journal of Mechanics - B/Fluids 27 (6) (2008) 643–664. doi:10.1016/j.euromechflu.2008.01.001

    work page doi:10.1016/j.euromechflu.2008.01.001 2008

  13. [13]

    M. A. Schalkers, M. Möller, Efficient and fail-safe quantum algorithm for the transport equation, Journal of Computational Physics 502 (2024) 112816. doi:10.1016/j.jcp.2024.112816

    work page doi:10.1016/j.jcp.2024.112816 2024

  14. [14]

    T. G. Draper, Addition on a quantum computer. URLhttp://arxiv.org/pdf/quant-ph/0008033

    work page internal anchor Pith review Pith/arXiv arXiv

  15. [15]

    Budinski, Quantum algorithm for the advection–diffusion equation simulated with the lattice boltzmann method, Quantum Information Processing 20 (2) (2021)

    L. Budinski, Quantum algorithm for the advection–diffusion equation simulated with the lattice boltzmann method, Quantum Information Processing 20 (2) (2021). doi:10.1007/s11128-021-02996-3

    work page doi:10.1007/s11128-021-02996-3 2021

  16. [16]

    L. Budinski, Quantum algorithm for the navier–stokes equations by us- ing the streamfunction-vorticity formulation and the lattice boltzmann method, International Journal of Quantum Information 20 (02) (2022). doi:10.1142/S0219749921500398. URLhttp://arxiv.org/pdf/2103.03804

    work page doi:10.1142/s0219749921500398 2022

  17. [17]

    Wawrzyniak, J

    D. Wawrzyniak, J. Winter, S. Schmidt, T. Indinger, C. F. Janßen, U. Schramm, N. A. Adams, Linearized quantum lattice- boltzmann method for the advection-diffusion equation using dy- namic circuits, Computer Physics Communications 317 (2025) 109856. doi:10.1016/j.cpc.2025.109856

    work page doi:10.1016/j.cpc.2025.109856 2025

  18. [18]

    Sanavio, S

    C. Sanavio, S. Succi, Lattice boltzmann–carleman quantum algorithm and circuit for fluid flows at moderate reynolds number, AVS Quantum Science 6 (2) (2024). doi:10.1116/5.0195549. URLhttp://arxiv.org/pdf/2310.17973

    work page doi:10.1116/5.0195549 2024

  19. [19]

    Carleman, Application de la théorie des équations intégrales linéaires aux systèmes d’équations différentielles non linéaires, Acta Mathematica 59 (0) (1932) 63–87

    T. Carleman, Application de la théorie des équations intégrales linéaires aux systèmes d’équations différentielles non linéaires, Acta Mathematica 59 (0) (1932) 63–87. doi:10.1007/BF02546499

    work page doi:10.1007/bf02546499 1932

  20. [20]

    Itani, Towards a quantum algorithm for lattice boltzmann (qalb) simulation with a nonlinear collision term, Dissertation, New York Uni- versity, NY, USA (03//2025)

    W. Itani, Towards a quantum algorithm for lattice boltzmann (qalb) simulation with a nonlinear collision term, Dissertation, New York Uni- versity, NY, USA (03//2025). 38

    2025

  21. [21]

    Lăcătuş, M

    M. Lăcătuş, M. Möller, Surrogate quantum circuit design for the lattice boltzmann collision operator. doi:10.48550/arXiv.2507.12256

    work page doi:10.48550/arxiv.2507.12256

  22. [22]

    Quantum computing in the NISQ era and beyond,

    J. Preskill, Quantum computing in the nisq era and beyond, Quantum 2 (2018) 79. doi:10.22331/q-2018-08-06-79

    work page internal anchor Pith review doi:10.22331/q-2018-08-06-79 2018

  23. [23]

    Cerezo, A

    M. Cerezo, A. Arrasmith, R. Babbush, S. C. Benjamin, S. Endo, K. Fu- jii, J. R. McClean, K. Mitarai, X. Yuan, L. Cincio, P. J. Coles, Varia- tional quantum algorithms, Nature Reviews Physics 3 (9) (2021) 625–

    2021

  24. [24]

    doi:10.1038/s42254-021-00348-9

    work page doi:10.1038/s42254-021-00348-9

  25. [25]

    Boltzmann, Weitere studien über das wärmegleichgewicht unter gas- molekülen, in: Sitzungsberichte der Kaiserlichen Akademie der Wis- senschaften, Vol

    L. Boltzmann, Weitere studien über das wärmegleichgewicht unter gas- molekülen, in: Sitzungsberichte der Kaiserlichen Akademie der Wis- senschaften, Vol. 66, Carl Gerold’s Sohn, Wien, 1872, pp. 275–370

  26. [26]

    Chapman, T

    S. Chapman, T. G. Cowling, The mathematical theory of non-uniform gases: An account of the kinetic theory of viscosity, thermal conduction, and diffusion in gases, 3rd Edition, Cambridge mathematical library, Cambridge University Press, Cambridge and New York, 1990

    1990

  27. [27]

    Abe, Derivation of the lattice boltzmann method by means of the discrete ordinate method for the boltzmann equation, Journal of Com- putational Physics 131 (1) (1997) 241–246

    T. Abe, Derivation of the lattice boltzmann method by means of the discrete ordinate method for the boltzmann equation, Journal of Com- putational Physics 131 (1) (1997) 241–246. doi:10.1006/jcph.1996.5595

    work page doi:10.1006/jcph.1996.5595 1997

  28. [28]

    He, L.-S

    X. He, L.-S. Luo, Theory of the lattice boltzmann method: From the boltzmann equation to the lattice boltzmann equation, Physical Review E 56 (6) (1997) 6811–6817. doi:10.1103/PhysRevE.56.6811

    work page doi:10.1103/physreve.56.6811 1997

  29. [29]

    G. R. McNamara, G. Zanetti, Use of the boltzmann equation to simulate lattice gas automata, Physical review letters 61 (20) (1988) 2332–2335. doi:10.1103/PhysRevLett.61.2332

    work page doi:10.1103/physrevlett.61.2332 1988

  30. [30]

    Succi, R

    S. Succi, R. Benzi, F. Higuera, The lattice boltzmann equation: A new toolforcomputationalfluid-dynamics, PhysicaD:NonlinearPhenomena 47 (1-2) (1991) 219–230. doi:10.1016/0167-2789(91)90292-H

    work page doi:10.1016/0167-2789(91)90292-h 1991

  31. [31]

    A. A. Mohamad, Lattice Boltzmann Method, Springer London, London,

  32. [32]

    doi:10.1007/978-0-85729-455-5

    work page doi:10.1007/978-0-85729-455-5

  33. [33]

    Krüger, H

    T. Krüger, H. Kusumaatmaja, A. Kuzmin, O. Shardt, G. Silva, E. M. Viggen, The Lattice Boltzmann Method, Springer International Pub- lishing, Cham, 2017. doi:10.1007/978-3-319-44649-3. 39

    work page doi:10.1007/978-3-319-44649-3 2017

  34. [34]

    S. Succi, The lattice Boltzmann equation for fluid dynamics and be- yond, Numerical mathematics and scientific computation, Clarendon Press and Oxford University Press, Oxford and New York, 2001

    2001

  35. [35]

    P. L. Bhatnagar, E. P. Gross, M. Krook, A model for collision pro- cesses in gases. i. small amplitude processes in charged and neu- tral one-component systems, Physical Review 94 (3) (1954) 511–525. doi:10.1103/PhysRev.94.511

    work page doi:10.1103/physrev.94.511 1954

  36. [36]

    Quantum circuit learning,

    K.Mitarai, M.Negoro, M.Kitagawa, K.Fujii, Quantumcircuitlearning, Physical Review A 98 (3) (2018). doi:10.1103/PhysRevA.98.032309

    work page doi:10.1103/physreva.98.032309 2018

  37. [37]

    M. A. Nielsen, I. L. Chuang, Quantum computation and quantum in- formation, 10th Edition, Cambridge University Press, Cambridge and New York, 2010. doi:10.1017/CBO9780511976667

    work page doi:10.1017/cbo9780511976667 2010

  38. [38]

    Schuld, R

    M. Schuld, R. Sweke, J. J. Meyer, Effect of data encoding on the expres- sive power of variational quantum-machine-learning models, Physical Review A 103 (3) (2021). doi:10.1103/PhysRevA.103.032430

    work page doi:10.1103/physreva.103.032430 2021

  39. [39]

    Ostaszewski, E

    M. Ostaszewski, E. Grant, M. Benedetti, Structure optimization for parameterizedquantumcircuits, Quantum5(2021)391. doi:10.22331/q- 2021-01-28-391

    work page doi:10.22331/q- 2021

  40. [40]

    F. J. Gil Vidal, D. O. Theis, Input redundancy for parameterized quan- tum circuits, Frontiers in Physics 8 (2020). doi:10.3389/fphy.2020.00297

    work page doi:10.3389/fphy.2020.00297 2020

  41. [41]

    Pérez-Salinas, A

    A. Pérez-Salinas, A. Cervera-Lierta, E. Gil-Fuster, J. I. Latorre, Data re-uploading for a universal quantum classifier, Quantum 4 (2020) 226. doi:10.22331/q-2020-02-06-226

    work page doi:10.22331/q-2020-02-06-226 2020

  42. [42]

    Casas, A

    B. Casas, A. Cervera-Lierta, Multidimensional fourier series with quantum circuits, Physical Review A 107 (6) (2023). doi:10.1103/PhysRevA.107.062612. URLhttp://arxiv.org/pdf/2302.03389

    work page doi:10.1103/physreva.107.062612 2023

  43. [43]

    Landman, S

    J. Landman, S. Thabet, C. Dalyac, H. Mhiri, E. Kashefi, Classically ap- proximating variational quantum machine learning with random fourier features. doi:10.48550/arXiv.2210.13200. 40

    work page doi:10.48550/arxiv.2210.13200

  44. [44]

    S. Shin, Y. S. Teo, H. Jeong, Exponential data encoding for quantum supervised learning, Physical Review A 107 (1) (2023). doi:10.1103/PhysRevA.107.012422

    work page doi:10.1103/physreva.107.012422 2023

  45. [45]

    Mhiri, L

    H. Mhiri, L. Monbroussou, M. Herrero-Gonzalez, S. Thabet, E. Kashefi, J. Landman, Constrained and vanishing expressivity of quantum fourier models, Quantum 9 (2025) 1847. doi:10.22331/q-2025-09-03-1847

    work page doi:10.22331/q-2025-09-03-1847 2025

  46. [46]

    Wiedmann, M

    M. Wiedmann, M. Periyasamy, D. D. Scherer, Fourier anal- ysis of variational quantum circuits for supervised learning. doi:10.48550/arXiv.2411.03450

    work page doi:10.48550/arxiv.2411.03450

  47. [47]

    J. R. McClean, S. Boixo, V. N. Smelyanskiy, R. Babbush, H. Neven, Barren plateaus in quantum neural network training landscapes, Nature communications 9 (1) (2018) 4812. doi:10.1038/s41467-018-07090-4

    work page doi:10.1038/s41467-018-07090-4 2018

  48. [48]

    Ortiz Marrero, M

    C. Ortiz Marrero, M. Kieferová, N. Wiebe, Entanglement- induced barren plateaus, PRX Quantum 2 (4) (2021). doi:10.1103/PRXQuantum.2.040316

    work page doi:10.1103/prxquantum.2.040316 2021

  49. [49]

    Holmes, A

    Z. Holmes, A. Arrasmith, B. Yan, P. J. Coles, A. Albrecht, A. T. Sorn- borger, Barren plateaus preclude learning scramblers, Physical review letters 126 (19) (2021) 190501. doi:10.1103/PhysRevLett.126.190501

    work page doi:10.1103/physrevlett.126.190501 2021

  50. [50]

    Holmes, K

    Z. Holmes, K. Sharma, M. Cerezo, P. J. Coles, Connecting ansatz ex- pressibility to gradient magnitudes and barren plateaus, PRX Quantum 3 (1) (2022). doi:10.1103/PRXQuantum.3.010313

    work page doi:10.1103/prxquantum.3.010313 2022

  51. [51]

    URLhttps://pennylane.ai/

    Pennylane: Quantum programming software (03/01/2025). URLhttps://pennylane.ai/

    2025

  52. [52]

    Corbetta, A

    A. Corbetta, A. Gabbana, V. Gyrya, D. Livescu, J. Prins, F. Toschi, To- ward learning lattice boltzmann collision operators, The European phys- ical journal. E, Soft matter 46 (3) (2023) 10. doi:10.1140/epje/s10189- 023-00267-w

    work page doi:10.1140/epje/s10189- 2023

  53. [53]

    S. Sim, P. D. Johnson, A. Aspuru-Guzik, Expressibility and entan- gling capability of parameterized quantum circuits for hybrid quantum– classical algorithms, Advanced Quantum Technologies 2 (12) (2019). doi:10.1002/qute.201900070. 41

    work page doi:10.1002/qute.201900070 2019

  54. [54]

    On Information and Sufficiency,

    S. Kullback, R. A. Leibler, On information and sufficiency, The Annals of Mathematical Statistics 22 (1) (1951) 79–86. doi:10.1214/aoms/1177729694

    work page doi:10.1214/aoms/1177729694 1951

  55. [55]

    Życzkowski, H.-J

    K. Życzkowski, H.-J. Sommers, Average fidelity between ran- dom quantum states, Physical Review A 71 (3) (2005). doi:10.1103/PhysRevA.71.032313

    work page doi:10.1103/physreva.71.032313 2005

  56. [56]

    D. A. Meyer, N. R. Wallach, Global entanglement in multiparticle systems, Journal of Mathematical Physics 43 (9) (2002) 4273–4278. doi:10.1063/1.1497700

    work page doi:10.1063/1.1497700 2002

  57. [57]

    A. J. Scott, C. M. Caves, Entangling power of the quantum baker s map, Journal of Physics A: Mathematical and General 36 (36) (2003) 9553–9576. doi:10.1088/0305-4470/36/36/308

    work page doi:10.1088/0305-4470/36/36/308 2003

  58. [58]

    The power of quantum neural networks,

    A. Abbas, D. Sutter, C. Zoufal, A. Lucchi, A. Figalli, S. Woerner, The power of quantum neural networks, Nature computational science 1 (6) (2021) 403–409. doi:10.1038/s43588-021-00084-1

    work page doi:10.1038/s43588-021-00084-1 2021

  59. [59]

    Abbas, D

    A. Abbas, D. Sutter, A. Figalli, S. Woerner, Effective dimension of ma- chine learning models

  60. [60]

    doi:10.48550/arXiv.2001.10872

    O.Berezniuk, A.Figalli, R.Ghigliazza, K.Musaelian, Ascale-dependent notion of effective dimension. doi:10.48550/arXiv.2001.10872

    work page doi:10.48550/arxiv.2001.10872 2001

  61. [61]

    Ragone, B

    M. Ragone, B. N. Bakalov, F. Sauvage, A. F. Kemper, C. Ortiz Marrero, M. Larocca, M. Cerezo, A lie algebraic theory of barren plateaus for deep parameterized quantum circuits, Nature communications 15 (1) (2024)

    2024

  62. [62]

    doi:10.1038/s41467-024-49909-3

    work page doi:10.1038/s41467-024-49909-3

  63. [63]

    M. E. Brachet, D. Meiron, S. Orszag, B. Nickel, R. Morf, U. Frisch, The taylor-green vortex and fully developed turbulence, Journal of Statistical Physics 34 (5-6) (1984) 1049–1063. doi:10.1007/BF01009458. 42

    work page doi:10.1007/bf01009458 1984