pith. sign in

arxiv: 2604.27135 · v1 · submitted 2026-04-29 · 🪐 quant-ph

Parametrized Variational Quantum Tomography

V. A. Penas , M. Losada , D. Tielas , F. Holik This is my paper

Pith reviewed 2026-05-07 10:34 UTC · model grok-4.3

classification 🪐 quant-ph
keywords quantum state tomographyvariational methodsmaximum entropydensity matrix reconstructionparametrized cost functionquantum state estimationincomplete measurements
0
0 comments X

The pith

A parametrized cost function interpolates between norms to unify variational quantum tomography methods and achieve higher fidelity to maximum-entropy states.

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

Quantum state tomography reconstructs quantum states from measurement data, but incomplete data allows many possible density matrices. The paper introduces a parametrized cost function that blends the 1-norm and infinity-norm approaches, connecting standard variational quantum tomography with its infinity variant. By adjusting the parameters, users can steer the reconstructed state toward the maximum-entropy solution in a controlled manner. This yields better agreement with the maximum-entropy estimator than the previous infinity-norm method while remaining computationally feasible. The approach addresses the underdetermined nature of the problem by offering tunable exploration of compatible states.

Core claim

The central claim is that a single parametrized cost function, which interpolates between the 1-norm and the infinity norm, unifies VQT and VQT∞ and permits controlled navigation among density matrices consistent with the data, producing reconstructions with higher fidelity to the MaxEnt solution than VQT∞ alone.

What carries the argument

A parametrized cost function that interpolates between the 1-norm and the infinity norm, allowing tunable selection among compatible density matrices.

If this is right

  • Reconstructions can be tuned to balance different norms and achieve closer agreement with maximum-entropy states.
  • The method preserves computational tractability while improving fidelity.
  • Users gain a way to explore the space of possible quantum states compatible with incomplete measurements.
  • Standard VQT and VQT∞ become special cases within one framework.

Where Pith is reading between the lines

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

  • Such tunable methods could extend to other underdetermined inverse problems in quantum information where multiple estimators exist.
  • The interpolation parameter might be optimized adaptively based on the specific measurement setup.
  • Testing on noisy data or larger systems would reveal whether the fidelity gains remain stable.

Load-bearing premise

The assumption that the parametrized cost function can be optimized efficiently without introducing instabilities and that higher fidelity to the maximum-entropy state is inherently preferable.

What would settle it

Apply the parametrized method to a set of incomplete quantum measurements on a known state, compute the fidelity to the maximum-entropy reconstruction, and check if it exceeds that of VQT∞ for some parameter values while matching the data.

Figures

Figures reproduced from arXiv: 2604.27135 by D. Tielas, F. Holik, M. Losada, V. A. Penas.

Figure 1
Figure 1. Figure 1: FIG. 1: Average trace distance between target and view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2: Average fidelity between target and view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3: (a) Histograms of fidelities between VQT view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4: Average fidelity between target and view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5: (a) Histograms of fidelities between PVQT with view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7: Average Kullback-Leibler divergence between view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9: Average fidelity between different VQTs and view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8: Average Von Neumann entropy for different view at source ↗
Figure 10
Figure 10. Figure 10: FIG. 10: Average fidelity between different VQTs and view at source ↗
Figure 12
Figure 12. Figure 12: FIG. 12: Average fidelity between different VQTs and view at source ↗
Figure 11
Figure 11. Figure 11: FIG. 11: Average fidelity between different VQTs and view at source ↗
read the original abstract

Quantum state tomography provides a fundamental framework for reconstructing quantum states. When the measurement data are not informationally complete, the observed statistics admit multiple compatible density matrices, making the reconstruction problem inherently underdetermined and calling for the selection of a meaningful estimator. Two well-established approaches to address this ambiguity are Maximum Entropy (MaxEnt) and Variational Quantum Tomography (VQT). A variant of VQT, named VQT$_\infty$, has been introduced to reproduce MaxEnt-like solutions. In this work, we generalize this approach by introducing a parametrized cost function that interpolates between the 1-norm and the infinity norm, thereby unifying VQT and VQT$_\infty$ within a single framework. By tuning the associated hyperparameters, the proposed method enables controlled exploration of the set of compatible density matrices. We show that this interplay yields reconstructed states with higher fidelity to the MaxEnt solution than those obtained via VQT$_\infty$ while preserving computational tractability.

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

0 major / 3 minor

Summary. The paper introduces a parametrized cost function for variational quantum tomography (VQT) that interpolates between the 1-norm (corresponding to standard VQT) and the infinity norm (VQT∞). This unifies the two approaches within a single framework controlled by a tunable hyperparameter. The central claim is that appropriate tuning of this parameter enables controlled exploration of the set of density matrices compatible with incomplete measurement data, yielding estimators with higher fidelity to the maximum-entropy (MaxEnt) solution than VQT∞ while preserving computational tractability. The manuscript presents the mathematical formulation, discusses the interpolation properties, and provides numerical demonstrations on example quantum states.

Significance. If the central claims hold, the work supplies a flexible, single-parameter method for selecting among underdetermined reconstructions in quantum state tomography, directly bridging VQT and MaxEnt estimators. This is useful for experiments with incomplete data, where practitioners often prefer MaxEnt-like solutions for their information-theoretic properties. The unification and the reported fidelity gains constitute a practical advance; the emphasis on retained tractability is a strength for near-term implementations.

minor comments (3)
  1. [§3.1, Eq. (7)] §3.1, Eq. (7): the interpolation is defined via a convex combination of norms, but the text does not explicitly state the admissible range of the hyperparameter λ (or equivalent) or prove that the resulting cost remains convex for all λ in that range; adding this would clarify the optimization landscape.
  2. [§4.3, Figure 4] §4.3, Figure 4: the fidelity curves are shown for a fixed number of shots; it would be helpful to include error bars from multiple random initializations or to report the variance across hyperparameter sweeps to substantiate the claim of consistent improvement.
  3. [Abstract and §1] The abstract and §1 state that the method 'preserves computational tractability,' yet no explicit scaling with system size or comparison of iteration counts versus VQT∞ is provided; a brief complexity remark or timing table would strengthen this assertion.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive assessment of our work and the recommendation for minor revision. The referee's summary correctly captures the introduction of the parametrized cost function that interpolates between the 1-norm and infinity norm, unifying VQT and VQT∞ while enabling higher fidelity to MaxEnt reconstructions. As no specific major comments were listed in the report, we have no individual points requiring detailed rebuttal or clarification at this stage.

Circularity Check

0 steps flagged

No significant circularity in the derivation chain

full rationale

The paper introduces a parametrized cost function interpolating between the 1-norm and infinity norm as a generalization of VQT and VQT∞, enabling hyperparameter-tuned exploration of compatible density matrices and claiming higher MaxEnt fidelity. No load-bearing derivation step reduces by construction to a fitted input, self-definition, or self-citation chain; the central construction is an independent ansatz presented as a unification framework rather than a renaming or forced equivalence. The abstract and described claims remain self-contained against external benchmarks such as standard VQT and MaxEnt without internal reduction to prior fitted values or author-overlapping uniqueness theorems.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the standard framework of quantum mechanics, the definition of compatible states, and the prior VQT method. The new element is the parametrization, which introduces one free hyperparameter.

free parameters (1)
  • interpolation hyperparameter
    A tunable parameter that controls the blend between 1-norm and infinity-norm in the cost function, chosen to optimize fidelity to MaxEnt.
axioms (1)
  • domain assumption The space of density matrices compatible with the measurement data is non-empty and convex.
    Standard assumption in quantum state tomography when data is underdetermined.

pith-pipeline@v0.9.0 · 5468 in / 1364 out tokens · 66044 ms · 2026-05-07T10:34:14.998245+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

28 extracted references · 28 canonical work pages

  1. [1]

    Nielsen and I

    M. Nielsen and I. Chuang,Quantum Computation and Quantum Information (Cambridge Series on Information and the Natural Sciences), 1st ed. (Cambridge University Press, 2004)

    work page 2004

  2. [2]

    Evangelista, C

    T. Evangelista, C. Lavor, and W. R. M. Rabelo, Int. J. Mod. Phys. C22, 95 (2011)

    work page 2011

  3. [3]

    T. O. Maciel and R. O. Vianna, Quantum Inf. Comput. 12, 0442 (2012)

    work page 2012

  4. [4]

    Quantum tomography,

    G. Mauro D’Ariano, M. G. Paris, and M. F. Sacchi, “Quantum tomography,” inAdvances in Imaging and Electron Physics(Elsevier, 2003) p. 205–308

    work page 2003

  5. [5]

    M. G. A. Paris and J. ˇReh´ aˇ cek, eds.,Quantum State Es- timation, Lecture Notes in Physics, Vol. 649 (Springer, Berlin, Heidelberg, 2004)

    work page 2004

  6. [6]

    Plenio, Steven T

    M. Cramer, M. B. Plenio, S. T. Flammia, R. Somma, D. Gross, S. D. Bartlett, O. Landon-Cardinal, D. Poulin, and Y.-K. Liu, Nature Communications1(2010), 10.1038/ncomms1147

    work page doi:10.1038/ncomms1147 2010

  7. [7]

    Gross, Y.-K

    D. Gross, Y.-K. Liu, S. T. Flammia, S. Becker, and J. Eisert, Physical Review Letters105, 150401 (2010)

    work page 2010

  8. [8]

    S. T. Flammia, D. Gross, Y.-K. Liu, and J. Eisert, New Journal of Physics14, 095022 (2012)

    work page 2012

  9. [9]

    Efficient state estimation on quantum processors,

    V. G. Avella, A. V. Vargas, T. M. Vergara, K. de la Ossa Doria, J. Czartowski, D. Main, G. Araneda, A. Del- gado, and D. Goyeneche, “Efficient state estimation on quantum processors,” (2025), arXiv:2510.16117 [quant- ph]

    work page arXiv 2025

  10. [10]

    Bendersky, F

    A. Bendersky, F. Pastawski, and J. P. Paz, Physical Review Letters100, 190403 (2008)

    work page 2008

  11. [11]

    C. T. Schmiegelow, A. Bendersky, M. A. Larotonda, and J. P. Paz, Physical Review Letters107, 100502 (2011)

    work page 2011

  12. [12]

    Aaronson, SIAM Journal on Computing49, STOC18 (2020), preliminary version in FOCS 2018

    S. Aaronson, SIAM Journal on Computing49, STOC18 (2020), preliminary version in FOCS 2018

    work page 2020

  13. [13]

    Huang, R

    H.-Y. Huang, R. Kueng, and J. Preskill, Nature Physics 16, 1050 (2020)

    work page 2020

  14. [14]

    T´ oth, W

    G. T´ oth, W. Wieczorek, D. Gross, R. Krischek, C. Schwemmer, and H. Weinfurter, Phys. Rev. Lett.105, 250403 (2010)

    work page 2010

  15. [15]

    Corte, M

    I. Corte, M. Losada, D. Tielas, F. Holik, and L. Reb´ on, Physica A: Statistical Mechanics and its Applications 611, 128427 (2023)

    work page 2023

  16. [16]

    Losada, F

    M. Losada, F. Holik, C. Massri, and A. Plastino, Quan- tum Information Processing18, 293 (2019)

    work page 2019

  17. [17]

    Tielas, M

    D. Tielas, M. Losada, L. Reb´ on, and F. Holik, Quantum Information Processing21, 227 (2022)

    work page 2022

  18. [18]

    Holik, M

    F. Holik, M. Losada, G. Zerr, L. Reb´ on, and D. Tielas, Quantum Information Processing24, 330 (2025)

    work page 2025

  19. [19]

    T. O. Maciel, A. T. Ces´ ario, and R. O. Vianna, Int. J. Mod. Phys. C22, 1361 (2011)

    work page 2011

  20. [20]

    de Klerk,Aspects of Semidefinite Programming: Inte- rior Point Algorithms and Selected Applications(Kluwer Academic Publishers, 2002)

    E. de Klerk,Aspects of Semidefinite Programming: Inte- rior Point Algorithms and Selected Applications(Kluwer Academic Publishers, 2002)

    work page 2002

  21. [21]

    Boyd and L

    S. Boyd and L. Vandenberghe,Convex Optimization (Cambridge University Press, 2004)

    work page 2004

  22. [22]

    D. S. Gon¸ calves, C. Lavor, M. A. Gomes-Ruggiero, A. T. Ces´ ario, R. O. Vianna, and T. O. Maciel, Physical Re- view A87(2013), 10.1103/physreva.87.052140

    work page doi:10.1103/physreva.87.052140 2013

  23. [23]

    Buzek, G

    V. Buzek, G. Drobny, R. Derka, G. Adam, and H. Wiedemann, Chaos, Solitons & Fractals10, 981 (1999)

    work page 1999

  24. [24]

    Helmberg, F

    C. Helmberg, F. Rendl, R. Vanderbei, and H. Wolkowicz, SIAM J. Optim.6, 342 (1996)

    work page 1996

  25. [25]

    Nemirovski and M

    A. Nemirovski and M. Todd, Acta Numer.17, 191 (2008)

    work page 2008

  26. [26]

    Życzkowski, K

    K. Zyczkowski, K. A. Penson, I. Nechita, and B. Collins, J. Math. Phys.52, 062201 (2011), 10.1063/1.3595693

    work page doi:10.1063/1.3595693 2011

  27. [27]

    J. M. Renes, R. Blume-Kohout, A. J. Scott, and C. M. Caves, J. Math. Phys.45, 2171 (2004)

    work page 2004

  28. [28]

    J. M. Renes, R. Blume-Kohout, A. J. Scott, and C. M. Caves, Journal of Mathematical Physics 45, 2171 (2004), https://pubs.aip.org/aip/jmp/article pdf/45/6/2171/19274161/21711online.pdf

    work page 2004