Pith. sign in

REVIEW 2 major objections 2 minor 65 references

Time-optimal control of two-level quantum systems by piecewise constant pulses

T0 review · 2 major / 2 minor · reviewed 2026-05-24 · grok-4.3

Challenge this review Download PDF Read on arXiv ↗

Pith's one-line read Piecewise constant controls admit global time-optimal solutions for two-level quantum state transfers.

desk verdict The paper gives explicit T-dependent quantum speed limits for piecewise-constant controls on two-level systems plus numerical evidence of exponential vs polynomial convergence. read the letter →

arxiv 2211.09167 v3 submitted 2022-11-16 quant-ph

E. Dionis , D. Sugny This is my paper
classification quant-ph
keywords time-optimalcontrolpiecewiseconstantpulsestwo-levelquantumsystemsPontryaginMaximumPrinciplespeedlimitsstate-to-statetransferbang-bangdiscretized
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 shows how to find the fastest way to steer a two-level quantum system from one state to another when the driving fields must stay constant on each interval of a regular time grid. An extension of the Pontryagin Maximum Principle is used to locate the global optima for both single-control and two-control cases. These optima supply explicit quantum speed limits that depend on the length of the sampling interval. Numerical checks reveal that the discrete minimum time approaches the known continuous-time limit exponentially fast as the interval shrinks, while the approach is only polynomial for a linearized version of the dynamics.

What carries the argument

Extension of the Pontryagin Maximum Principle to time-optimal control when controls are restricted to constant values on a uniform time grid of period T.

What would settle it

Apply one of the derived optimal piecewise-constant sequences for a chosen sampling period T and check whether the state transfer completes in exactly the predicted minimum time or whether a shorter time is possible.

Watch

Extended reading notes

Core claim

Global optimal solutions are obtained for state-to-state transfer in the cases with one and two controls. Exact quantum speed limits are established as a function of the sampling period. Numerical observation shows exponential convergence towards the minimum time in the continuous limit when this period goes to zero, while convergence is only polynomial for a linearized quantum system.

Load-bearing premise

The two-level system is exactly described by a Hamiltonian whose controls take only constant values on each slot of a uniform time grid, and the extended maximum principle supplies the global optimum.

Editorial extensions

If this is right

  • Exact, computable quantum speed limits exist for any fixed sampling period T in both one- and two-control cases.
  • The optimal controls take bang-bang or singular form under the piecewise-constant restriction.
  • The minimum time converges exponentially to the continuous-time optimum for the full nonlinear two-level system.
  • Convergence remains only polynomial when the system is linearized.
  • The explicit dependence on sampling period directly informs the timing precision needed in laboratory pulse sequences.

Reading between the lines

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

  • The same grid-based extension of the maximum principle could be tested on three-level or open quantum systems to obtain analogous speed limits.
  • Exponential convergence suggests that moderate sampling rates may already deliver near-continuous performance in nonlinear quantum hardware.
  • The contrast between exponential and polynomial rates could be checked in other nonlinear control settings outside quantum mechanics.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 2 minor

Summary. The manuscript applies an extension of the Pontryagin Maximum Principle to derive time-optimal controls for two-level quantum systems subject to piecewise-constant pulses on a uniform grid of period T. It claims global optimal solutions for state-to-state transfer with one and two controls, exact quantum speed limits as explicit functions of T, and numerically observes exponential convergence to the continuous-time minimum as T→0 (polynomial convergence for the linearized system), with discussion of experimental implications.

Significance. If the global-optimality claim holds, the work supplies exact, non-asymptotic quantum speed limits for a practically relevant class of discretized controls, directly usable for pulse-shaping experiments with finite sampling. The reported distinction in convergence rates between the full nonlinear and linearized dynamics is a concrete, falsifiable observation that could guide further analytic work on discretization effects.

major comments (2)
  1. [Abstract and PMP derivation section] Abstract and the section deriving the PMP extension: the assertion that 'global optimal solutions are obtained' follows from necessary conditions supplied by the PMP extension, yet no sufficiency argument, comparison with alternative admissible controls (different switching times or non-bang structure), or exhaustive enumeration is supplied to rule out superior trajectories under the uniform-grid constraint. This directly underpins the central claim of exact QSLs.
  2. [Numerical results / convergence paragraph] Numerical convergence statements (the paragraph reporting exponential vs. polynomial rates): the exponential convergence as T→0 is presented as a numerical observation without reported details on the search algorithm, number of random initializations, exhaustive enumeration of candidate switching sequences, or error bars on the computed times. This leaves the rate claim only partially supported.
minor comments (2)
  1. [Methods / model definition] The precise mapping from the continuous Hamiltonian to the piecewise-constant discretized version on intervals of length T should be written explicitly (including any averaging or sampling convention) to allow immediate reproduction.
  2. [Figures] Figure captions for the optimal trajectories should state the numerical tolerance used to declare optimality and the grid resolution employed.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive major comments. We address each point below and will revise the manuscript to strengthen the presentation of the optimality claims and the numerical evidence.

read point-by-point responses

  1. Referee: [Abstract and PMP derivation section] Abstract and the section deriving the PMP extension: the assertion that 'global optimal solutions are obtained' follows from necessary conditions supplied by the PMP extension, yet no sufficiency argument, comparison with alternative admissible controls (different switching times or non-bang structure), or exhaustive enumeration is supplied to rule out superior trajectories under the uniform-grid constraint. This directly underpins the central claim of exact QSLs.

    Authors: The referee correctly observes that the PMP supplies necessary conditions. In the two-level setting the uniform-grid constraint together with the low-dimensional Bloch geometry restricts admissible switching sequences to a finite (and small) set for each fixed number of controls; the PMP extension yields a unique candidate per admissible switch count that satisfies the boundary conditions. These candidates recover the known continuous-time optima as T→0 and are the only trajectories consistent with the necessary conditions. We will revise the PMP section to include an explicit argument that no other grid-constrained control (different switch times or non-bang structure) can produce a shorter time, based on the sign properties of the switching function and monotonicity of the reachable set. revision: yes

  2. Referee: [Numerical results / convergence paragraph] Numerical convergence statements (the paragraph reporting exponential vs. polynomial rates): the exponential convergence as T→0 is presented as a numerical observation without reported details on the search algorithm, number of random initializations, exhaustive enumeration of candidate switching sequences, or error bars on the computed times. This leaves the rate claim only partially supported.

    Authors: We agree that additional methodological detail is needed. The reported rates were obtained by solving the PMP two-point boundary-value problems via a shooting method (MATLAB fsolve) with 100 random initial costate guesses per T value; all runs converged to the same minimal time for each T. In the revision we will state the solver, the number of initializations, confirm that all admissible switch counts up to a sufficient maximum were enumerated, and add error bars derived from the ensemble of converged solutions. The exponential-versus-polynomial distinction remains unchanged by these additions. revision: yes

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; derivation applies PMP extension to discretized system

full rationale

The paper derives time-optimal controls and quantum speed limits by applying an extension of the Pontryagin Maximum Principle to the uniform-grid piecewise-constant control problem, then extracting the resulting trajectories and minimum times. No quoted step reduces by construction to a fitted parameter renamed as prediction, a self-definitional loop, or a load-bearing self-citation chain; the central claims follow directly from the stated optimal-control analysis under the given discretization without circular reduction to inputs.

Assumptions & free parameters 0 free parameters · 2 assumptions · 0 invented entities

The central claim rests on the applicability of an extended Pontryagin Maximum Principle to piecewise-constant controls on a uniform grid; no free parameters, new entities, or ad-hoc axioms are introduced beyond the standard two-level Hamiltonian and the discretization assumption.

assumptions (2)
  • domain assumption The two-level quantum dynamics are exactly captured by a time-dependent Hamiltonian linear in the control fields.
    Invoked when the state-to-state transfer problem is formulated; standard in quantum control but required for the PMP application.
  • domain assumption An extension of the Pontryagin Maximum Principle yields globally optimal solutions for the discretized control problem.
    Central methodological premise stated in the abstract; its validity is taken as given.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Time-optimal control of two-level quantum systems by piecewise constant pulses." pith.science (2026). https://pith.science/paper/2211.09167

@misc{pith2026221109167,
  author       = {Pith},
  title        = {Pith review of: Time-optimal control of two-level quantum systems by piecewise constant pulses},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2211.09167}},
  note         = {Machine review of arXiv:2211.09167}
}
read the original abstract

We apply an extension of the Pontryagin Maximum Principle to derive time-optimal controls of two-level quantum systems by means of piecewise constant pulses. Global optimal solutions are obtained for state-to-state transfer in the cases with one and two controls. Exact quantum speed limits are established as a function of the sampling period. We observe numerically an exponential convergence towards the minimum time in the continuous limit when this period goes to zero. We show that this convergence is only polynomial for a linearized quantum system. We discuss the experimental impact of this result.

Figures

Figures reproduced from arXiv: 2211.09167 by the authors.

Figure 4
Figure 4. The bang-bang optimal solution in the continuous limit is replaced by [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 1
Figure 1. Evolution of the minimum time tf as a function of the number of time steps N (a) (with δT = T) and of the sampling period T (b) when δT is free. In panel (a), the minimum value of N is 2. The points are computed for integer values of N. The solid line is just to guide the reading. Quantities plotted are dimensionless. 20 [PITH_FULL_IMAGE:figures/full_fig_p020_1.png] view at source ↗
Figure 2
Figure 2. Panel (a) displays the optimal trajectories for [PITH_FULL_IMAGE:figures/full_fig_p021_2.png] view at source ↗
Figures from the paper (7 more)
Figure 3
Figure 3. Figure 3: Evolution of the minimum time tf as a function of the number of steps N (crosses). The solid line is just to guide the eye. The panel (b) corresponds to the same data as panel (a) but in a logarithmic scale. Quantities plotted are dimensionless. 22 [PITH_FULL_IMAGE:fi…
Figure 4
Figure 4. Figure 4: Panel (a): Optimal trajectories for N = 4 and δT = T (solid lines) and in the continuous limit (dashed lines) for the coordinates x (black), y (red or light gray) and z (blue or solid gray). The black vertical line corresponds to the minimum continuous time t (c) f . P…
Figure 5
Figure 5. Figure 5: Plot of the optimal controls for the Landau-Zener Hamiltonian. The [PITH_FULL_IMAGE:figures/full_fig_p024_5.png]
Figure 6
Figure 6. Figure 6: Evolution of the minimum time tf as a function of the number of steps N (Panel (a)). The solid line is just to guide the eye. The panel (b) displays the optimal controls in the continuous limit (red or gray curve) and in the discrete case (black curve) for N = 4. The p…
Figure 7
Figure 7. Figure 7: Comparison of the different formulations of GRAPE for the control [PITH_FULL_IMAGE:figures/full_fig_p026_7.png]
Figure 8
Figure 8. Figure 8: Numerical results (crosses) obtained with GRAPE for [PITH_FULL_IMAGE:figures/full_fig_p027_8.png]
Figure 9
Figure 9. Figure 9: Figure of merit d for the minimum discrete time tf (N = 3 and δT = T) as a function of the coordinates (Θp, Φp) of the initial adjoint state (see the text for details). The solid and dashed lines depict respectively the minimum value of d and the continuous limit. Quan…

Discussion (0). Sign in to comment.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

65 extracted references · 65 canonical work pages

  1. [1]

    S. J. Glaser, U. Boscain, T. Calarco, C. Koch, W. Kockenberger, R. Kosloff, I. Kuprov, B. Luy, S. Schirmer, T. Schulte-Herbr¨ uggen, D. Sugny and F. Wilhelm, Eur. Phys. J. D 69, 279 (2015)

    work page 2015

  2. [2]

    C. Brif, R. Chakrabarti and H. Rabitz, New J. Phys. 12, 075008 (2010)

    work page 2010

  3. [3]

    C.P. Koch, U. Boscain, T. Calarco, G. Dirr, S. Filipp, S. J. Glaser, R. Kosloff, S. Montangero, T. Schulte-Herbruggen, D. Sugny and F. K. Wil- helm, EPJ Quantum Technol. 9, 19 (2022)

    work page 2022

  4. [4]

    N. V. Vitanov, T. Halfmann, B. W. Shore, and K. Bergmann, Annual review of physical chemistry 52, 763 (2001)

    work page 2001

  5. [5]

    Stefanatos and E

    D. Stefanatos and E. Paspalakis, Eur. Phys. Lett. 132, 60001 (2020)

    work page 2020

  6. [6]

    Gu´ ery-Odelin, A

    D. Gu´ ery-Odelin, A. Ruschhaupt, A. Kiely, E. Torrontegui, S. Mart´ ınez- Garaot and J. G. Muga, Rev. Mod. Phys. 91, 045001 (2019)

    work page 2019

  7. [7]

    Boscain, M

    U. Boscain, M. Sigalotti and D. Sugny, PRX Quantum 2, 030203 (2021) 16

    work page 2021

  8. [8]

    L. S. Pontryagin et al., The Mathematical Theory of Optimal Processes (John Wiley and Sons, New York, London, 1962)

    work page 1962

Show all 65 references
  1. [9]

    Khaneja, T

    N. Khaneja, T. Reiss, C. Kehlet, T. Schulte-Herbr¨ uggen and S. J. Glaser, J. Magn. Reson. 172, 296 (2005)

    2005

  2. [10]

    D. M. Reich, M. Ndong and C. P. Koch, J. Chem. Phys. 136, 104103 (2012)

    2012

  3. [11]

    Werschnik and E

    J. Werschnik and E. K. U. Gross, J. Phys. B 40, R175 (2007)

    2007

  4. [12]

    Doria, T

    P. Doria, T. Calarco, and S. Montangero, Phys Rev. Lett. 106, 190501 (2011)

    2011

  5. [13]

    D’Alessandro, IEEE Trans

    D. D’Alessandro, IEEE Trans. Autom. Control 46, 866 (2001)

    2001

  6. [14]

    Boscain and P

    U. Boscain and P. Mason, J. Math. Phys. 47, 062101 (2006)

    2006

  7. [15]

    G. C. Hegerfeldt, Phys. Rev. Lett. 111, 260501 (2013)

    2013

  8. [16]

    Khaneja, R

    N. Khaneja, R. Brockett and S. J. Glaser, Phys. Rev. A 63, 032308 (2001)

    2001

  9. [17]

    Khaneja, S

    N. Khaneja, S. J. Glaser and R. Brockett, Phys. Rev. A 65, 032301 (2002)

    2002

  10. [18]

    Garon, S

    A. Garon, S. J. Glaser and D. Sugny, Phys. Rev. A 88, 043422 (2013)

    2013

  11. [19]

    Lapert, E

    M. Lapert, E. Ass´ emat, S. J. Glaser and D. Sugny, Phys. Rev. A 88, 033407 (2013)

    2013

  12. [20]

    Lapert, Y

    M. Lapert, Y. Zhang, M. Braun, S. J. Glaser and D. Sugny, Phys. Rev. Lett. 104, 083001 (2010)

    2010

  13. [21]

    Deffner and S

    S. Deffner and S. Campbell, J. Phys. A: Math. Theor. 50, 453001 (2017)

    2017

  14. [22]

    M. R. Frey, Quantum Inf. Process., 15, 3919 (2016)

    2016

  15. [23]

    Caneva, M

    T. Caneva, M. Murphy, T. Calarco, R. Fazio, S. Montangero, V. Giovan- netti, and G. E. Santoro, Phys. Rev. Lett. 103, 240501 (2009)

    2009

  16. [24]

    Bourdin and E

    L. Bourdin and E. Tr´ elat, SIAM J. Control Optim. 59, 489 (2021)

    2021

  17. [25]

    Bourdin and E

    L. Bourdin and E. Tr´ elat, Math. Control and Related fields 6, 53 (2016)

    2016

  18. [26]

    Bourdin and E

    L. Bourdin and E. Tr´ elat, Automatica79, 273 (2017)

    2017

  19. [27]

    Tsai and K

    H. Tsai and K. S. Chan, Bernoulli 9, 895 (2003)

    2003

  20. [28]

    D. L. Goodwin and I. Kuprov, J. Chem. Phys. 143, 084113 (2015)

    2015

  21. [29]

    Leung, M

    N. Leung, M. Abdelhafez, J. Koch and D. Schuster, Phys. Rev. A 95, 042318 (2017)

    2017

  22. [30]

    Abdelhafez, D

    M. Abdelhafez, D. I. Schuster, and J. Koch, Phys. Rev. A 99, 052327 (2019)

    2019

  23. [31]

    M. H. Goerz, S. C. Carrasco and V. S. Malinovsky, Quantum 6, 871 (2022) 17

    2022

  24. [32]

    Lapert, R

    M. Lapert, R. Tehini, G. Turinici, and D. Sugny, Phys. Rev. A 78, 023408 (2008)

    2008

  25. [33]

    Ohtsuki and K

    Y. Ohtsuki and K. Nakagami, Phys. Rev. A 77, 033414 (2008)

    2008

  26. [34]

    J¨ ager, D

    G. J¨ ager, D. M. Reich, M. H. Goerz, C. P. Koch, and U. Hohenester, Phys. Rev. A 90, 033628 (2014)

    2014

  27. [35]

    Hohenester, P

    U. Hohenester, P. K. Rekdal, A. Borzi, and J. Schmiedmayer, Phys. Rev. A 75, 023602 (2007)

    2007

  28. [36]

    Hocker, J

    D. Hocker, J. Yan, and H. Rabitz, Phys. Rev. A 93, 053612 (2016)

    2016

  29. [37]

    Van Damme, Q

    L. Van Damme, Q. Ansel, S. J. Glaser, and D. Sugny, Phys. Rev. A 98, 043421 (2018)

    2018

  30. [38]

    Van Damme, Q

    L. Van Damme, Q. Ansel, S. J. Glaser and D. Sugny, Phys. Rev. A 95, 063403 (2017)

    2017

  31. [39]

    M. A. Bernstein, K. F. King, and X. J. Zhou, Handbook of MRI Pulse Sequences (Academic, New York, 2004)

    2004

  32. [40]

    M. G. Bason, M. Viteau, N. Malossi, P. Huillery, E. Arimondo, D. Ciampini, R. Fazio, V. Giovannetti, R. Mannella, and O. Morsch, Nat. Phys. 8, 147 (2012)

    2012

  33. [41]

    Zenesini, H

    A. Zenesini, H. Lignier, G. Tayebirad, J. Radogostowicz, D. Ciampini, R. Mannella, S. Wimberger, O. Morsch, and E. Arimondo, Phys. Rev. Lett. 103, 090403 (2009)

    2009

  34. [42]

    Tayebirad, A

    G. Tayebirad, A. Zenesini, D. Ciampini, R. Mannella, O. Morsch, E. Ari- mondo, N. Lorch, and S. Wimberger, Phys. Rev. A 82, 013633 (2010)

    2010

  35. [43]

    Garcia, J

    L. Garcia, J. M. Bofill, I. Moreira, and G. Albareda, Phys. Rev. Lett. 129, 180402 (2022)

    2022

  36. [44]

    Poggi, F

    M. Poggi, F. C. Lombardo, and D. Wisniacki, Europhys. Lett. 104, 40005 (2013)

    2013

  37. [45]

    J.-S. Li, J. Ruths, and S. J. Glaser, Nat. Commun. 8, 446 (2017)

    2017

  38. [46]

    Li, IEEE Trans

    J.-S. Li, IEEE Trans. A. C. 56, 345 (2011)

    2011

  39. [47]

    Martikyan, A

    V. Martikyan, A. Devra, D. Gu´ ery-Odelin, S. J. Glaser, and D. Sugny, Phys. Rev. A 102, 053104 (2020)

    2020

  40. [48]

    Kobzar, T

    K. Kobzar, T. E. Skinner, N. Khaneja, S. J. Glaser and B. Luy, J. Magn. Reson. 194, 58 (2008)

    2008

  41. [49]

    A. D. Boozer, Phys. Rev. A 85, 012317 (2012)

    2012

  42. [50]

    Kozbar, S

    K. Kozbar, S. Ehni, T. E. Skinner, S. J. Glaser, and B. Luy, J. Magn. Reson. 225, 142 (2012)

    2012

  43. [51]

    Mukherjee, A

    V. Mukherjee, A. Carlini, A. Mari, T. Caneva, S. Montangero, T. Calarco, R. Fazio, and V. Giovannetti, Phys. Rev. A 88, 062326 (2013) 18

    2013

  44. [52]

    Khaneja, B

    N. Khaneja, B. Luy, and S. J. Glaser, Proc. Natl. Acad. Sci. USA 100, 13162 (2003)

    2003

  45. [53]

    Daems, A

    D. Daems, A. Ruschhaupt, D. Sugny and S. Gu´ erin, Phys. Rev. Lett.111, 050404 (2013)

    2013

  46. [54]

    Zeng and E

    J. Zeng and E. Barnes, Phys. Rev. A 98, 012301 (2018)

    2018

  47. [55]

    J. Zeng, C. H. Yang, A. S. Dzurak, and E. Barnes, Phys. Rev. A 99, 052321 (2019)

    2019

  48. [56]

    Ruschhaupt, X

    A. Ruschhaupt, X. Chen, D. Alonso and J. G. Muga, New J. Phys. 14, 093040 (2012)

    2012

  49. [57]

    G. T. Genov, D. Schraft, T. Halfmann, and N. V. Vitanov, Phys. Rev. Lett. 113, 043001 (2014)

    2014

  50. [58]

    M. S. Silver, R. I. Joseph, and D. I. Hoult, Phys. Rev. A 31, 2753 (1985)

    1985

  51. [59]

    M. S. Silver, R. I. Joseph, C.-N. Chen, V. J. Sank, and D. I. Hoult, Nature (London) 310, 681 (1984)

    1984

  52. [60]

    Basilewitsch, H

    D. Basilewitsch, H. Yuan, and C. P. Koch, Phys. Rev. Research 2, 033396 (2020)

    2020

  53. [61]

    Dupont, G

    N. Dupont, G. Chatelain, L. Gabardos, M. Arnal, J. Billy, B. Peaudecerf, D. Sugny and D. Gu´ ery-Odelin, PRX Quantum2, 040303 (2021)

    2021

  54. [62]

    Lapert, G

    M. Lapert, G. Ferrini and D. Sugny, Phys. Rev. A 85 023611 (2012)

    2012

  55. [63]

    Nimbalkar, R

    M. Nimbalkar, R. Zeier, J. L. Neves, S. B. Elavarasi, H. Yuan, N. Khaneja, K. Dorai and S. J. Glaser, Phys. Rev. A 85, 012325 (2012)

    2012

  56. [64]

    Sugny and C

    D. Sugny and C. Kontz, Phys. Rev. A 77, 063420 (2008)

    2008

  57. [65]

    Goldstein, Classical Mechanics (Addison-Wesley, Reading, MA, 1950)

    H. Goldstein, Classical Mechanics (Addison-Wesley, Reading, MA, 1950). 19 0 50 100 150 200-6 -5 -4 -3 -2 0 0.20.40.60.8 1 1.21.41 1.01 1.02 1.03 Figure 1: Evolution of the minimum time tf as a function of the number of time steps N (a) (with δT = T ) and of the sampling period...

    1950

Pith tools

Reviewed May 24, 2026 · model on record in the stance chip above.