arxiv: 2603.23656 · v2 · submitted 2026-03-24 · 🪐 quant-ph · cond-mat.stat-mech· hep-th· nucl-th

Information-Geometric Quantum Process Tomography of Single Qubit Systems

T. Koide , A. van de Venn This is my paper

Pith reviewed 2026-05-15 00:14 UTC · model grok-4.3

classification 🪐 quant-ph cond-mat.stat-mechhep-thnucl-th

keywords information geometryquantum process tomographysingle qubitquantum exponential familyGKSL master equationlinear regressionmixed statesthermodynamic speed limits

0 comments

The pith

For single qubits an information-geometric inequality saturates to an exact equality, turning continuous-time process tomography into ordinary linear regression.

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

The paper derives an information-geometric inequality that holds for quantum states evolving under arbitrary dynamics, whether Markovian or non-Markovian. In the single-qubit case the inequality becomes an equality because the density operator lies in the quantum exponential family with the Pauli matrices as sufficient statistics. This exact identity supplies a closed-form linear estimator that recovers the parameters of the underlying dynamics directly from observed trajectories. The estimator avoids the local-minima traps of nonlinear optimization routines and is demonstrated on the Gorini-Kossakowski-Sudarshan-Lindblad master equation. Numerical tests confirm reliable performance except near the pure-state boundary, where the inverse metric diverges and error mitigation becomes necessary.

Core claim

We establish an exact information-geometric inequality that remains valid regardless of the underlying dynamics, encompassing both Markovian and non-Markovian evolutions within the mixed-state domain. For single qubits this inequality saturates into a strict equality because the density matrix belongs to the quantum exponential family with the Pauli matrices serving as sufficient statistics. The resulting identity enables a non-iterative linear regression approach to continuous-time quantum process tomography that bypasses the local minima issues common in non-linear optimization. The method is used to estimate the Hamiltonian and dissipation parameters of the Gorini-Kossakowski-Sudarshan-Ln

What carries the argument

The information-geometric inequality that saturates to equality for single qubits because the density matrix belongs to the quantum exponential family with Pauli matrices as sufficient statistics.

If this is right

The equality supplies a non-iterative linear regression procedure for continuous-time quantum process tomography.
The procedure avoids local-minima problems that arise in nonlinear optimization methods.
Hamiltonian and dissipation parameters of the GKSL master equation can be recovered directly from observed trajectories.
Error mitigation is required near the pure-state boundary where the inverse metric becomes singular.

Where Pith is reading between the lines

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

The same saturation mechanism may appear in any low-dimensional system whose state manifold admits a finite sufficient-statistic representation.
The linear estimator could be combined with existing filtering techniques to enable real-time parameter tracking during experiments.
The geometric distance underlying the equality provides a natural figure of merit for comparing different tomography protocols on single qubits.

Load-bearing premise

The single-qubit density matrix belongs to the quantum exponential family with the Pauli matrices serving as sufficient statistics.

What would settle it

A set of single-qubit density-matrix trajectories in which the linear-regression estimates of the GKSL parameters deviate systematically from the values recovered by independent nonlinear fitting would falsify the claimed saturation to equality.

Figures

Figures reproduced from arXiv: 2603.23656 by A. van de Venn, T. Koide.

**Figure 2.** Figure 2: FIG. 2. Convergence of the Hamiltonian parameter estimation. The estimated Hamiltonian [PITH_FULL_IMAGE:figures/full_fig_p017_2.png] view at source ↗

**Figure 3.** Figure 3: FIG. 3. Convergence of the dissipation parameter estimation. The estimated dissipation pa [PITH_FULL_IMAGE:figures/full_fig_p018_3.png] view at source ↗

read the original abstract

We establish an exact information-geometric inequality that remains valid regardless of the underlying dynamics, encompassing both Markovian and non-Markovian evolutions within the mixed-state domain. This inequality can be viewed as an extension of thermodynamic speed limits, which are typically formulated as inequalities. For single qubits, we show that this inequality saturates into a strict equality because the density matrix belongs to the quantum exponential family with the Pauli matrices serving as sufficient statistics. From a practical perspective, this identity enables a non-iterative linear regression approach to continuous-time quantum process tomography, bypassing the local minima issues common in non-linear optimization. We demonstrate the efficiency of this method by estimating the Hamiltonian and dissipation parameters of the Gorini-Kossakowski-Sudarshan-Lindblad (GKSL) master equation. Numerical simulations confirm the validity of this geometric estimator and highlight the necessity of error mitigation near the pure-state boundary where the inverse metric becomes singular.

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.

Desk Editor's Note private letter to a colleague

This paper turns an info-geometric inequality into a linear estimator for single-qubit GKSL tomography by using the exponential-family structure of qubits.

read the letter

The main takeaway is that for single qubits the authors show an information-geometric inequality saturates to an exact equality because any qubit state belongs to the quantum exponential family generated by the Pauli matrices. This identity replaces nonlinear optimization with ordinary linear regression on the Hamiltonian and dissipation parameters of the GKSL equation, which is the practical contribution. They run numerical simulations that support the estimator and correctly flag the singularity of the inverse metric near pure states as a point needing error mitigation. That combination of a clean derivation step and an implementable workaround is the part that stands out. The work stays within its stated scope and does not claim generality beyond single qubits. The soft spots are modest but real. Everything rests on the single-qubit exponential-family property, which is standard but also caps how far the method travels. The abstract and stress-test give no sign of circularity or invented entities, yet the full error analysis and data-handling rules are not visible here, so a referee would need to check those details carefully. The simulations look adequate on the surface but would benefit from explicit tests of numerical stability when the metric inverse is ill-conditioned. This is aimed at experimental groups doing single-qubit device characterization or theorists working on open-system parameter estimation. A reader who needs a fast, non-iterative alternative to standard tomography will get immediate value; someone looking for higher-dimensional extensions will not. I would send it to peer review. The core logic is straightforward, the limitation is acknowledged, and the practical payoff is clear enough to justify referee time.

Referee Report

2 major / 3 minor

Summary. The manuscript establishes an information-geometric inequality valid for arbitrary (Markovian or non-Markovian) dynamics on mixed states. For single-qubit systems it saturates to an exact equality because every qubit density matrix lies in the quantum exponential family generated by the three Pauli matrices as sufficient statistics. The resulting identity is used to replace nonlinear optimization with a non-iterative linear regression estimator for the Hamiltonian and dissipation parameters of the GKSL master equation; numerical simulations are presented to support the method, with a note on error mitigation near the pure-state boundary where the inverse metric is singular.

Significance. If the saturation holds, the work supplies a parameter-free, linear-regression route to continuous-time single-qubit process tomography that avoids local-minima problems of conventional fitting. The approach rests on a standard structural property of qubit states and therefore inherits the reproducibility of that property; it also supplies a concrete, falsifiable prediction (exact equality) that can be checked on any single-qubit experiment. The restriction to qubits and the need for singularity handling are explicitly acknowledged.

major comments (2)

[Abstract and information-geometry derivation] The central saturation claim is asserted in the abstract and presumably derived in the information-geometry section, yet the explicit reduction of the general inequality to an identity via the exponential-family property (i.e., the explicit form of the divergence or Fisher metric that cancels) is not shown in sufficient algebraic detail. A short derivation that starts from ρ = (I + r·σ)/2 and arrives at the equality would make the load-bearing step verifiable.
[Linear-regression estimator and numerical simulations] The linear-regression estimator for the GKSL parameters is presented as bypassing nonlinear optimization, but the manuscript does not specify the precise design matrix or the handling of the singular inverse metric in the regression; without these, it is unclear whether the reported numerical success is robust or depends on ad-hoc data exclusion near pure states.

minor comments (3)

[Notation] Notation for the information-geometric quantities (e.g., the precise definition of the divergence appearing in the inequality) should be introduced once and used consistently; several symbols appear without prior definition in the abstract.
[Figures] Figure captions for the simulation results should state the number of trajectories, the noise model, and the precise metric used for the regression residual so that the plots can be reproduced from the text alone.
[Numerical results] A brief comparison with at least one standard iterative QPT method (e.g., maximum-likelihood or Bayesian) on the same simulated data would strengthen the practical-advantage claim.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the positive assessment and for recommending minor revision. The comments are constructive and help strengthen the clarity of the central derivation and the practical implementation of the estimator. We address each major comment below and will revise the manuscript accordingly.

read point-by-point responses

Referee: [Abstract and information-geometry derivation] The central saturation claim is asserted in the abstract and presumably derived in the information-geometry section, yet the explicit reduction of the general inequality to an identity via the exponential-family property (i.e., the explicit form of the divergence or Fisher metric that cancels) is not shown in sufficient algebraic detail. A short derivation that starts from ρ = (I + r·σ)/2 and arrives at the equality would make the load-bearing step verifiable.

Authors: We agree that an explicit algebraic derivation of the saturation would make the key step more verifiable. In the revised manuscript we will insert a short, self-contained derivation in the information-geometry section. Beginning from the Bloch representation ρ = (I + r · σ)/2, we will show how the relevant quantum divergence (or Fisher metric) reduces exactly to an identity because the three Pauli matrices constitute a complete set of sufficient statistics for the qubit exponential family. The cancellation that converts the general inequality into equality will be written out step by step. revision: yes
Referee: [Linear-regression estimator and numerical simulations] The linear-regression estimator for the GKSL parameters is presented as bypassing nonlinear optimization, but the manuscript does not specify the precise design matrix or the handling of the singular inverse metric in the regression; without these, it is unclear whether the reported numerical success is robust or depends on ad-hoc data exclusion near pure states.

Authors: We acknowledge that the precise design matrix and the concrete procedure for treating the singular inverse metric were not stated explicitly. In the revision we will (i) give the explicit form of the design matrix that appears in the linear regression for the GKSL coefficients and (ii) describe the error-mitigation strategy, including any regularization or threshold-based handling of data points near the pure-state boundary. These additions will demonstrate that the reported numerical performance rests on a reproducible, non-ad-hoc procedure. revision: yes

Circularity Check

0 steps flagged

No significant circularity; saturation follows from standard qubit exponential-family structure

full rationale

The paper first establishes an information-geometric inequality that holds for arbitrary dynamics (Markovian or non-Markovian) in the mixed-state regime. Saturation to equality is then invoked only for single qubits because every qubit density matrix lies in the quantum exponential family generated by the three Pauli matrices; this is a pre-existing structural fact (ρ = (I + r·σ)/2 admits an exponential representation with those generators) and is not derived from or defined by the inequality itself. The resulting identity is used to replace nonlinear optimization with linear regression on GKSL parameters. No step reduces a prediction to a fitted input by construction, no uniqueness theorem is smuggled via self-citation, and the central derivation remains independent of the tomography application. The single-qubit restriction and metric singularity near pure states are explicitly flagged, confirming the logic is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the standard property that single-qubit states form a quantum exponential family; no free parameters or new entities are introduced in the abstract.

axioms (1)

domain assumption Single-qubit density matrices belong to the quantum exponential family with Pauli matrices as sufficient statistics.
Invoked to establish saturation of the inequality to equality for qubits.

pith-pipeline@v0.9.0 · 5464 in / 1158 out tokens · 43903 ms · 2026-05-15T00:14:50.246297+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

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

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel; dAlembert_cosh_solution_aczel echoes

?

echoes
ECHOES: this paper passage has the same mathematical shape or conceptual pattern as the Recognition theorem, but is not a direct formal dependency.

the density matrix belongs to the quantum exponential family with the Pauli matrices serving as sufficient statistics... the BKM metric corresponds to the Hessian of the exponential family potential
IndisputableMonolith/Foundation/CostAlphaLog.lean costAlphaLog_high_calibrated_iff echoes

?

echoes
ECHOES: this paper passage has the same mathematical shape or conceptual pattern as the Recognition theorem, but is not a direct formal dependency.

gμν(θ) := ⟨∂μĤ, ∂νĤ⟩cc ... gμν(θ) = Ξμν(θ)

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

43 extracted references · 43 canonical work pages

[1]

M. A. Nielsen and I. L. Chuang,Quantum Computation and Quantum Information(Cam- bridge, Cambridge Univ. Press, 2000)

work page 2000
[2]

Hayashi,Quantum Information: An Introduction(Berlin Heidelberg, Springer-Verlag, 2006)

M. Hayashi,Quantum Information: An Introduction(Berlin Heidelberg, Springer-Verlag, 2006)

work page 2006
[3]

M. M. Wild,Quantum Information Theory(Cambridge, Cambridge Univ. Press, 2017)

work page 2017
[4]

Watrous,The Theory of Quantum Information(Cambridge, Cambridge Univ

J. Watrous,The Theory of Quantum Information(Cambridge, Cambridge Univ. Press, 2018)

work page 2018
[5]

Amari,Differential-Geometrical Methods in Statistics(Springer-Verlag, Berlin, 1985)

S. Amari,Differential-Geometrical Methods in Statistics(Springer-Verlag, Berlin, 1985)

work page 1985
[6]

Amari and H

S. Amari and H. Nagaoka,Methods of Information Geometry(Oxford University Press, Ox- ford, 2000)

work page 2000
[7]

N. Ay, J. Jost, H. V. Lˆ e, and L. Schwachh¨ ofer,Information Geometry(Springer, Cham, 2017)

work page 2017
[8]

Stochastic Thermodynamic Interpretation of Information Geometry

S. Ito, “Stochastic Thermodynamic Interpretation of Information Geometry”, Phys. Rev. Lett. 121, 030605 (2018)

work page 2018
[9]

Speed Limit for Classical Stochastic Processes

N. Shiraishi, K. Funo, and K. Saito, “Speed Limit for Classical Stochastic Processes”, Phys. Rev. Lett.121, 070601 (2018)

work page 2018
[10]

Speed limit for open quantum systems

K. Funo, N. Shiraishi, and K. Saito, “Speed limit for open quantum systems”, New J. Phys. 21, 013006 (2019)

work page 2019
[11]

Stochastic Time Evolution, Information Geometry, and the Cram´ er- Rao Bound

S. Ito and A. Dechant, “Stochastic Time Evolution, Information Geometry, and the Cram´ er- Rao Bound”, Phys. Rev. X10, 021056 (2020)

work page 2020
[12]

Time-information uncertainty relations in thermodynamics

S. B. Nicholson, L. P. Garc´ ıa-Pintos, A. del Campo, and J. R. Green, “Time-information uncertainty relations in thermodynamics”, Nat. Phys.16, 1211 (2020)

work page 2020
[13]

Quantum Thermodynamic Uncertainty Relation for Continuous Measure- ment

Y. Hasegawa, “Quantum Thermodynamic Uncertainty Relation for Continuous Measure- ment”, Phys. Rev. Lett.125, 050601 (2020). 21

work page 2020
[14]

Unified speed limits in classical and quantum dynamics via temporal Fisher information

T. Nishiyama and Y. Hasegawa, “Unified speed limits in classical and quantum dynamics via temporal Fisher information”, arXiv:2504.04790 (2025)

work page arXiv 2025
[15]

Thermodynamic Unification of Optimal Transport: Thermodynamic Uncertainty Relation, Minimum Dissipation, and Thermodynamic Speed Limits

T. Van Vu and K. Saito “Thermodynamic Unification of Optimal Transport: Thermodynamic Uncertainty Relation, Minimum Dissipation, and Thermodynamic Speed Limits”, Phys. Rev. X13, 011013 (2023)

work page 2023
[16]

Generalized Geometric Quantum Speed Limits

D. P. Pires, M. Cianciaruso, L. C. C´ eleri, G. Adesso, and D. O. Soares-Pinto, “Generalized Geometric Quantum Speed Limits”, Phys. Rev. X6, 021031 (2016)

work page 2016
[17]

Paris and J

M. Paris and J. Rehacek, eds.,Quantum State Estimation(Berlin Heidelberg, Springer-Verlag, 2004)

work page 2004
[18]

Continuous-variable optical quantum-state tomography

A. I. Lvovsky and M. G. Raymer, “Continuous-variable optical quantum-state tomography”, Rev. Mod. Phys.81, 299 (2009)

work page 2009
[19]

Maximum-likelihood estimation of quantum processes,

J. Fiur´ aˆ sek and Z. Hradil, “Maximum-likelihood estimation of quantum processes,” Phys. Rev. A,63, 020101 (2001)

work page 2001
[20]

Maximum-likelihood reconstruction of completely positive maps,

M. F. Sacchi, “Maximum-likelihood reconstruction of completely positive maps,” Phys. Rev. A63054104 (2001)

work page 2001
[21]

Realization of quantum process tomography in NMR,

A. M. Childs, I. L. Chuang, and D. W. Leung, “Realization of quantum process tomography in NMR,” Phys. Rev. A64, 012314 (2001)

work page 2001
[22]

Quantum inference of states and processes,

M. Jeˇ zek, J. Fiur´ aˇ sek, and Z. Hradil, “Quantum inference of states and processes,” Phys. Rev. A68, 012305 (2003)

work page 2003
[23]

Ancilla-Assisted Quantum Process Tomography

J. B. Altepeteret al., “Ancilla-Assisted Quantum Process Tomography”, Phys. Rev. Lett.90, 193601 (2003)

work page 2003
[24]

Quantum process tomography of a controlled-NOT gate,

J. L. O’Brien, G. J. Pryde, A. Gilchrist, D. F. V. James, N. K. Langford, T. C. Ralph, and A. G. White, “Quantum process tomography of a controlled-NOT gate,” Phys. Rev. Lett.93, 080502 (2004)

work page 2004
[25]

Process reconstruction: From unphysical to physical maps via maximum likelihood,

M. Ziman, M. Plesch, and V. Buˇ zek, “Process reconstruction: From unphysical to physical maps via maximum likelihood,” Phys. Rev. A72, 022106 (2005)

work page 2005
[26]

Quantum-process tomography: Resource analysis of different strategies,

M. Mohseni, A. T. Rezakhani, and D. A. Lidar, “Quantum-process tomography: Resource analysis of different strategies,” Phys. Rev. A77, 032322 (2008)

work page 2008
[27]

Maximum-likelihood coherent-state quantum process tomogra- phy,

A. Anis and A. I. Lvovsky, “Maximum-likelihood coherent-state quantum process tomogra- phy,” New J. Phys.14, 105021 (2012)

work page 2012
[28]

Self-consistent quantum process tomography,

S. T. Merkel, J. M. Gambetta, J. A. Smolin, S. Poletto, A. D. C´ orcoles, B. R. Johnson, C. 22 A. Ryan, and M. Steffen, “Self-consistent quantum process tomography,” Phys. Rev. A87, 062119 (2013)

work page 2013
[29]

Lindblad tomography of a superconducting quantum processor,

G. O. Samachet al., “Lindblad tomography of a superconducting quantum processor,” Phys. Rev. Applied18, 064056 (2022)

work page 2022
[30]

Quantum State Tomography via Linear Regression Estimation

B. Qi, Z. Hou, L. Li, D. Dong, G. Xiang and G. Guo, “Quantum State Tomography via Linear Regression Estimation”, Scientific Reports3, 3496 (2013)

work page 2013
[31]

Full reconstruction of a 14-qubit state within four hours,

Z. Hou, H.-S. Zhong, Y. Tian, D. Dong, B. Qi, L. Li, Y. Wang, F. Nori, G.-Y. Xiang, C.-F. Li, and G.-C. Guo, “Full reconstruction of a 14-qubit state within four hours,” New J. Phys. 18, 083036 (2016)

work page 2016
[32]

Adaptive quantum state tomography via linear regression estimation: Theory and two-qubit experiment,

B. Qi, Z. Hou, Y. Wang, D. Dong, H.-S. Zhong, L. Li, G.-Y. Xiang, H. M. Wiseman, C.-F. Li, and G.-C. Guo, “Adaptive quantum state tomography via linear regression estimation: Theory and two-qubit experiment,” npj Quantum Information3, 19 (2017)

work page 2017
[33]

A Quantum Hamilto- nian Identification Algorithm: Computational Complexity and Error Analysis

Y. Wang, D. Dong, B. Qi, J. Zhang, I. R. Petersen and H. Yonezawa, “A Quantum Hamilto- nian Identification Algorithm: Computational Complexity and Error Analysis”, IEEE Trans. Autom. Control.63, 1388 (2018)

work page 2018
[34]

Compressed-sensing Lindbladian quantum tomography with trapped ions

D. Dobrynin, L. Cardarelli, M. M¨ uller and A. Bermudez, “Compressed-sensing Lindbladian quantum tomography with trapped ions”, Quantum Science and Technology10, 045041 (2025)

work page 2025
[35]

Monotone Metrics on Matrix Spaces

D. Petz, “Monotone Metrics on Matrix Spaces”, Linear Algebra Appl.244, 81 (1996)

work page 1996
[36]

C. W. Helstrom,Quantum Detection and Estimation Theory(Academic Press, New York, 1976)

work page 1976
[37]

Transport, Collective Motion and Brownian Motion

H. Mori, “Transport, Collective Motion and Brownian Motion”, Prog. Theor. Phys.33, 423 (1965)

work page 1965
[38]

Thermodynamic length in open quantum systems

M. Scandi and M. Perarnau-Llobet, “Thermodynamic length in open quantum systems”, Quantum3, 197 (2019)

work page 2019
[39]

Geometric decomposition of entropy production in out-of- equilibrium systems

A. Dechant, S.-i. Sasa, and S. Ito, “Geometric decomposition of entropy production in out-of- equilibrium systems”, Phys. Rev. Research4, L012034 (2022)

work page 2022
[40]

The Gravitational Aspect of Information: The Physical Reality of Asymmetric “Distance

T. Koide and A. van de Venn, “The Gravitational Aspect of Information: The Physical Reality of Asymmetric “Distance””, arXiv:2510.22664

work page arXiv
[41]

Torsion-Induced Quantum Fluctuations in Metric-Affine Grav- ity using the Stochastic Variational Method

T. Koide and A. van de Venn, “Torsion-Induced Quantum Fluctuations in Metric-Affine Grav- ity using the Stochastic Variational Method”, Symmetry18, 525 (2026). 23

work page 2026
[42]

Biconnection gravity as a statistical manifold

D. Iosifidis and K. Pallikaris, “Biconnection gravity as a statistical manifold”, Phys. Rev. D 108, 044026 (2023)

work page 2023
[43]

Lindblad evolution as gradient flow

G. Kaplanek, A. Maloney, J. Pollack and D. VanAllen, “Lindblad evolution as gradient flow”, Phys. Rev. A112, 042220 (2025). 24

work page 2025