Hybrid Cram\'er-Rao bound for Quantum Bayes-Point Estimation with Nuisance Parameters

Jianchao Zhang; Jun Suzuki

REVIEW 2 major objections 6 minor 40 references

Hybrid Cram\'er-Rao bound for Quantum Bayes-Point Estimation with Nuisance Parameters

T0 review · 2 major / 6 minor · reviewed 2026-08-04 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read For any measurement and any locally unbiased estimator, the hybrid mean-squared error of the interest parameters is bounded below by the inverse of a prior-averaged Schur-complement quantum Fisher matrix, the hpQFIM.

desk verdict A useful new quantum hybrid CRB that is almost certainly correct, but the written proof of Theorem 1 has a load-bearing gap and the numerics gloss over the regularity conditions. read the letter →

arxiv 2510.16810 v1 pith:KGCPNDLK submitted 2025-10-19 quant-ph

Jianchao Zhang , Jun Suzuki This is my paper · ORCID

classification quant-ph

keywords quantumparameterestimationnuisanceparametershybridBayes-pointFisherinformationCramér-RaoboundSchurcomplementvanTreesinequalityqubitmetrology

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

This paper proposes a hybrid estimation framework for quantum metrology in which the parameters of interest are treated as fixed and non-random, while nuisance parameters are integrated out with a prior. Its central claim is a Cramér-Rao-type inequality: for any measurement (POVM) and any locally unbiased estimator of the interest parameters, the hybrid mean-squared-error matrix is bounded below by the inverse of a prior-averaged Schur-complement quantum Fisher matrix, which the paper calls the hpQFIM. If true, this gives metrologists a rigorous way to exploit partial prior knowledge of nuisance variables—such as calibration drifts or unknown loss—without giving up local precision on the quantity that actually matters. The paper also proves two-sided bracketing inequalities that make the bound easy to compute or approximate, and it works out qubit examples where frequentist estimation is impossible but hybrid estimation has a finite, closed-form floor.

What carries the argument

The hpQFIM: for SLD QFIM blocks J_{ab}, define J^{(π)}_{I|N}(θI) = Eπ[JII] - Eπ[JIN] (Eπ[JNN] + Jπ)^{-1} Eπ[JNI], where Jπ is the classical Fisher information of the nuisance prior. The proof uses a covariance inequality on the augmented joint distribution p(x, θN|θI) = π(θN)p(x|θI, θN), forming the augmented information matrix G(π) = Eπ[J] + diag(0, Jπ), then taking the (I,I) block of its inverse.

What would settle it

Take a qubit model with a uniform nuisance prior on a compact interval and compute the boundary terms in the integration-by-parts step explicitly. If any boundary term is nonzero for some POVM and locally unbiased estimator, the inequality V ⪰ hpQFIM^{-1} fails (or requires a modified hpQFIM with boundary corrections). Concretely, test a translation-family model with uniform prior on [0, 2π) for boundary contributions.

Watch

Extended reading notes

Core claim

The paper establishes a Cramér-Rao-type lower bound for quantum estimation when some parameters are treated as fixed (interest) and others are integrated out with a prior (nuisance). The central quantity is the hybrid partial quantum Fisher information matrix (hpQFIM): prior-average the QFIM blocks, add the prior's Fisher information to the nuisance block, then take the Schur complement. The theorem states that for any POVM and any locally unbiased estimator of the interest parameters, the hybrid MSE matrix satisfies V_{θI,π}(Π, θ̂I) ⪰ (J^{(π)}_{I|N}(θI))^{-1}. The paper also proves the hpQFIM is bracketed between simple averages: Eπ[JII] ⪰ J^{(π)}_{I|N} ⪰ Eπ[JI|N]. In the exactly solvable B

Load-bearing premise

The proof requires the nuisance prior to be twice continuously differentiable and to decay at the boundary so all integration-by-parts boundary terms vanish; the numerical examples instead use uniform priors on closed intervals that do not satisfy these conditions.

Editorial extensions

If this is right

If the bound holds, hybrid estimation gives a universal precision limit for any locally unbiased estimator of the interest parameters, mixing frequentist and Bayesian treatments in one inequality.
A measurement designed from the hpQFIM depends only on the nuisance prior, not on the unknown nuisance value, so no adaptation to the nuisance is needed.
In models where the frequentist partial information J_{I|N} vanishes (score directions collinear), the prior Fisher term Jπ restores a finite bound: J^{(π)}_{I|N} = (r^2 sin^2 ϕ Jπ)/(r^2 sin^2 ϕ + Jπ).
The bracketing inequalities provide a cheap way to sandwich the exact hybrid bound: average the partial information matrices or average the interest block alone, avoiding repeated matrix inversions.

Reading between the lines

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

The numerical examples use uniform priors on compact intervals, which do not satisfy the smoothness/decay condition assumed in the theorem's proof; a limiting or boundary-term analysis is needed to confirm the bound holds exactly there.
Because the proof only uses the QFIM's positive semidefiniteness and a covariance inequality, the same construction should extend to right-logarithmic-derivative quantum Fisher information and other score choices.
In the Bloch-radius model, the bound depends on the prior only through Eπ[r^2]; this suggests a calibration protocol where one estimates the prior second moment offline and then certifies direction-estimation precision without runtime knowledge of the radius.
The open 'full hybrid model' with four parameter classes could unify existing partial Bayesian and partial frequentist bounds; the present framework supplies the natural interpolation.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Quick take: the hybrid partial QFIM and the bracketing inequalities in Theorem 2 are genuine additions to the quantum estimation toolbox. The statement of Theorem 1 is very likely correct; the written proof in Appendix A is not.

The core idea is clean: nuisance parameters get a prior, interest parameters stay frequentist, and the hpQFIM is the Schur complement of the prior-averaged QFIM plus prior Fisher information. The two-sided bounds in Theorem 2 are proved correctly and are actually useful: the middle quantity is a pain to compute, so bracketing it between simpler averages is a practical contribution. The examples are honest illustrations, not curve fitting, and the extra-rotation model nicely shows how a prior regularizes an otherwise unidentifiable phase. The classical hybrid CRB background (Messer, Fortunati et al.) is the right literature, and the self-citations are on point.

Where it falls down: the proof of Theorem 1 in Appendix A introduces an estimator θ̂N for the nuisance parameter that never appears in the theorem, and then asserts E[(θ̂N−θN)∇_{θN} log(πp)] = I. That is a Bayesian unbiasedness condition for θ̂N, not a consequence of the local unbiasedness of θ̂I. The covariance inequality applied to the full vector therefore does not yield the desired block bound. This is not a minor typo; it is the load-bearing step.

The good news is that the theorem can almost certainly be rescued with the standard trick: apply the covariance inequality to f1 = θ̂I − θI alone, using the full score vector. Then T = [I, 0] (the θN block vanishes by integration by parts because θ̂I and θI do not depend on θN), and the (I,I) block of the inverse is exactly the hpQFIM inverse. The authors should replace Appendix A with this argument.

A secondary soft spot: the stated regularity assumptions require π to be twice differentiable with vanishing boundary terms, but the numerical examples use uniform priors on compact intervals. For flat priors the integration-by-parts boundary terms need a separate argument (periodicity or a limiting prior). This is minor relative to the proof gap, but a referee should ask for it.

Who is this for? People working on quantum metrology with nuisance parameters, especially those who want a hybrid Bayesian/frequentist treatment. It is a methods paper, not new physics, and the numerics are illustrative. It deserves a serious referee; I would not desk-reject it. But I would require a corrected proof of Theorem 1 before publication, and the examples should be reconciled with the regularity assumptions.

Referee Report

2 major / 6 minor

Summary. The paper introduces a hybrid estimation framework for quantum metrology in which the parameters of interest are treated as fixed (frequentist) and nuisance parameters are treated as random with a prior. The central object is the hybrid partial quantum Fisher information matrix (hpQFIM), defined by prior-averaging the nuisance block of the SLD QFIM and taking a Schur complement. Theorem 1 claims a Cramér–Rao-type lower bound: for any POVM and any locally unbiased interest-parameter estimator, the hybrid mean-squared-error matrix is bounded below by the inverse hpQFIM. Theorem 2 gives two-sided approximations to the hpQFIM. The paper also presents numerical qubit examples and one analytically solvable Bloch-radius model. The main result is a quantum analogue of the classical hybrid CRB, with the stated proof relying on a van-Trees-type covariance inequality.

Significance. If Theorem 1 is correct, the hpQFIM is a natural and useful quantity for quantum estimation problems with nuisance parameters: it provides a bound that depends only on the prior over the nuisance, not on its unknown true value, and the two-sided bounds in Theorem 2 are computationally convenient. The framework is a sensible extension of classical hybrid CRB ideas to the quantum setting, and the solvable qubit example cleanly illustrates the effect of prior averaging. However, the proof of Theorem 1 as written is not valid, and the numerical examples do not satisfy the stated regularity conditions. The core idea is plausible and repairable, but the paper requires substantial revision before the main claim is established.

major comments (2)

[Appendix A, proof of Theorem 1 (Eqs. (26)–(29))] The proof introduces an estimator θ̂N for the nuisance parameters, defines f2(x)=θ̂N(x)−θN, and then claims T=E[f gᵀ]=I. The assertion that E[(θ̂N−θN)∇θNᵀ]=I is a Bayesian unbiasedness condition for θ̂N, but θ̂N is never defined in the theorem or in Section 2.1, and its existence/unbiasedness does not follow from the stated local unbiasedness of θ̂I. Without T=I, the block covariance inequality F ⪰ T G⁻¹ Tᵀ does not imply the desired (I,I)-block bound. The theorem is likely true, however: applying the covariance inequality to f1=θ̂I−θI alone with the full score vector gives T11=I and T12=0 (under the stated boundary conditions), and then V_{I} ⪰ (G⁻¹)_{11} = (J_{I|N}^{(π)})⁻¹, which is exactly the hpQFIM bound. The proof should be rewritten along these lines.
[Section 2.1 (after Definition 2) and Section 3] Theorem 1 assumes that the nuisance prior π is twice continuously differentiable and decays so that all boundary terms vanish in integration by parts. The numerical examples in Section 3 use uniform priors on compact intervals: θN ∼ U[0,2π) (Figures 1 and 2) and θN ∼ U(0,1] (Figure 3). These priors are not twice continuously differentiable and do not decay at the boundaries; the paper does not explain how the theorem applies to them or why the boundary terms are negligible. If boundary terms do not vanish, the claimed bound can fail. The authors should either restrict the examples to smooth priors with a limit argument or compute the boundary contributions and show they are negligible for the specific models considered.

minor comments (6)

[Section 3.1 vs Section 3.3] The additional-sine model is stated in Section 3.1 with θN ∈ [0,2π), but Section 3.3 says θN ∼ Unif(0,1]. Figure 2's caption says θN ∼ U[0,2π). Please make the domain consistent.
[Figure 2 caption] The caption refers to the 'phase with extra rotation' model, but the plot is for the additional-sine model. Correct the caption.
[Appendix A, last displayed equation] The notation V_{θI,θN,π}(Π, θ̂I, θ̂N) is introduced even though θ̂N is not defined in Theorem 1. Remove or define this symbol.
[Abstract and Section 1.1] The abstract states that the hybrid approach improves over pure point estimation because the optimal measurement depends only on the prior of the nuisance. The paper does not construct or characterize optimal measurements; it only derives lower bounds. Please soften this claim or add a remark distinguishing the lower bound from achievability.
[Section 2.1, Definition 2] The regularity assumption on π is stated as 'twice continuously differentiable in θN and decays sufficiently fast at the boundary (or at infinity)'. For compact parameter spaces, this should be made explicit (e.g., support strictly inside the interval) so that the later use of uniform priors does not appear contradictory.
[Various equations] There are minor typos and notation inconsistencies, e.g., 'Eπ[JθN]' vs 'Eπ[J_NN]' in Appendix A, and the use of 'Jπ' without definition in Section 3.2. Please proofread carefully.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the hybrid CR bound is derived from the prior-averaged Fisher information via the standard van Trees covariance inequality; it is not fitted, renamed, or reduced to the authors' prior results.

full rationale

The paper's central claim, Theorem 1, is not circular: the hpQFIM is defined as a prior-averaged Schur complement, and the bound V ≥ (J_{I|N}^{(π)})^{-1} is proven from an independent covariance inequality using the prior-augmented information matrix G^{(π)}. The proof does not assume the desired inequality; it derives the (I,I)-block bound from T G^{-1} T^T with T claimed to be the identity. No fitted parameters enter the hQPFIM, and the numerical examples compute the bound from closed-form QFIM blocks and prescribed priors, so there is no fitted-input-called-prediction issue. The authors do cite their own related work (e.g., refs. [7,8,32,39]), but these citations provide background and motivation; the main theorem does not rest on any unpublished or self-referential uniqueness claim. The manifest proof gap in Appendix A — the estimator θ̂_N is introduced without definition and the claim T=I requires a Bayesian unbiasedness condition for θ̂_N that is not established — is a correctness/rigor issue, not circularity. Similarly, the use of uniform priors on compact domains where the stated boundary-decay regularity condition fails affects the applicability of the theorem to the examples, but it is not a circular step. The derivation chain, as written, does not reduce to its own inputs, so the appropriate circularity score is 0.

Assumptions & free parameters 0 free parameters · 5 assumptions · 1 invented entities

The central bound is derived, not fitted: no free parameters are tuned and no physical entities are postulated. The derivation relies on standard QFIM/van Trees machinery plus regularity/invertibility assumptions; the new hpQFIM is a mathematical definition rather than an empirical input.

assumptions (5)

domain assumption Van Trees regularity: π(θN) is twice continuously differentiable and decays at the boundary so boundary terms vanish; expectation and derivatives can be interchanged.
Stated after Definition 2 and used in Appendix A to prove T=I and compute G. Uniform compact priors in Section 3 violate it.
standard math SLD quantum Fisher information matrix dominates the classical Fisher information matrix of any POVM (Braunstein–Caves inequality).
Invoked in Appendix A after Eq. (27), cited [15].
standard math Covariance (matrix) inequality F ⪰ T G^{-1} T^T.
Core of Appendix A; cited [28] (van Trees).
domain assumption The QFIM blocks JNN (and Eπ[JNN]+Jπ) are invertible so the Schur complement and hpQFIM are well-defined.
Definition 3 and Theorem 1 require inverses; degenerate models need generalized inverses, not discussed.
domain assumption There exists an estimator θ̂I that is locally unbiased at θI for every θN (Eq. (3)).
The bound applies only to such estimators; if none exist, the theorem is vacuous.

invented entities (1)

Hybrid partial quantum Fisher information matrix (hpQFIM)
purpose: Defines the lower-bound information quantity in Theorem 1 and the bracketing in Theorem 2.
A new mathematical matrix, not a physical entity; no external falsifiable handle. It is definitionally constructed, so it carries no independent evidential weight.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Hybrid Cram\'er-Rao bound for Quantum Bayes-Point Estimation with Nuisance Parameters." pith.science (2026). https://pith.science/paper/KGCPNDLK

@misc{pith2026251016810,
  author       = {Pith},
  title        = {Pith review of: Hybrid Cram\'er-Rao bound for Quantum Bayes-Point Estimation with Nuisance Parameters},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KGCPNDLK}},
  note         = {Machine review of arXiv:2510.16810}
}

read the original abstract

We develop a hybrid framework for quantum parameter estimation in the presence of nuisance parameters. In this Bayes-point scheme, the parameters of interest are treated as fixed non-random parameters while nuisance parameters are integrated out with respect to a prior (random parameters). Within this setting, we introduce the hybrid partial quantum Fisher information matrix (hpQFIM), defined by prior-averaging the nuisance block of the QFIM and taking a Schur complement, and derive a corresponding Cram\'er-Rao-type lower bound on the hybrid risk. We establish structural properties of the hpQFIM, including inequalities that bracket it between computationally tractable surrogates, as well as limiting behaviors under extreme priors. Operationally, the hybrid approach improves over pure point estimation since the optimal measurement for the parameters of interest depends only on the prior distribution of the nuisance, rather than on its unknown value. We illustrate the framework with analytically solvable qubit models and numerical examples, clarifying how partial prior information on nuisance variables can be systematically exploited in quantum metrology.

Figures

Figures reproduced from arXiv: 2510.16810 by the authors.

**Figure 2.** Comparison of the hpQFIM and its two approximations for the model (phase with [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗

**Figure 3.** Comparison of the hpQFIM and its two approximations for the model (anisotropic [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

40 extracted references · 1 linked inside Pith

[1]

Quantum sensing,

C. L. Degen, F. Reinhard, and P. Cappellaro, “Quantum sensing,”Reviews of Modern Physics, vol. 89, no. 3, p. 035002, 2017

2017
[2]

Advances in quantum metrology,

V. Giovannetti, S. Lloyd, and L. Maccone, “Advances in quantum metrology,”Nature photonics, vol. 5, no. 4, p. 222, 2011

2011
[3]

Quantum metrology with nonclassical states of atomic ensembles,

L. Pezze, A. Smerzi, M. K. Oberthaler, R. Schmied, and P. Treutlein, “Quantum metrology with nonclassical states of atomic ensembles,”Reviews of Modern Physics, vol. 90, no. 3, p. 035005, 2018

2018
[4]

Quantum estimation for quantum technology,

M. G. A. Paris, “Quantum estimation for quantum technology,”International Journal of Quantum Information, vol. 7, pp. 125–137, 2009

2009
[5]

The elusive heisenberg limit in quantum-enhanced metrology,

R. Demkowicz-Dobrza´ nski, J. Ko lody´ nski, and M. Gut ¸˘ a, “The elusive heisenberg limit in quantum-enhanced metrology,”Nature communications, vol. 3, no. 1, p. 1063, 2012

2012
[6]

Critical quantum metrology in the non-linear quantum rabi model,

Z.-J. Ying, S. Felicetti, G. Liu, and D. Braak, “Critical quantum metrology in the non-linear quantum rabi model,”Entropy, vol. 24, no. 8, p. 1015, 2022

2022
[7]

Nuisance parameter problem in quantum estimation theory: Tradeoff relation and qubit examples,

J. Suzuki, “Nuisance parameter problem in quantum estimation theory: Tradeoff relation and qubit examples,”Journal of Physics A: Mathematical and Theoretical, vol. 53, no. 26, p. 264001, 2020

2020
[8]

Quantum state estimation with nuisance parameters,

J. Suzuki, Y. Yang, and M. Hayashi, “Quantum state estimation with nuisance parameters,” Journal of Physics A: Mathematical and Theoretical, vol. 53, no. 45, p. 453001, 2020. 18

2020

Show all 40 references

[9]

Quantum limits in optical interferometry,

R. Demkowicz-Dobrza´ nski, M. Jarzyna, and J. Ko lody´ nski, “Quantum limits in optical interferometry,”Progress in Optics, vol. 60, pp. 345–435, 2015

2015
[10]

Imaging with quantum states of light,

P.-A. Moreau, E. Toninelli, T. Gregory, and M. J. Padgett, “Imaging with quantum states of light,”Nature Reviews Physics, vol. 1, no. 6, pp. 367–380, 2019

2019
[11]

Frequency stability degradation of an oscillator slaved to a periodically interrogated atomic resonator,

G. Santarelli, C. Audoin, A. Makdissi, P. Laurent, G. J. Dick, and A. Clairon, “Frequency stability degradation of an oscillator slaved to a periodically interrogated atomic resonator,” IEEE transactions on ultrasonics, ferroelectrics, and frequency control, vol. 45, no. 4, pp...

2002
[12]

Compatibility in multiparameter quantum metrology,

S. Ragy, M. Jarzyna, and R. Demkowicz-Dobrza´ nski, “Compatibility in multiparameter quantum metrology,”Physical Review A, vol. 94, no. 5, p. 052108, 2016

2016
[13]

Fisher information in quantum statistics,

O. E. Barndorff-Nielsen and R. D. Gill, “Fisher information in quantum statistics,”Journal of Physics A: Mathematical and General, vol. 33, no. 24, p. 4481, 2000

2000
[14]

Frequentist and bayesian quantum phase estimation,

Y. Li, L. Pezz` e, M. Gessner, Z. Ren, W. Li, and A. Smerzi, “Frequentist and bayesian quantum phase estimation,”Entropy, vol. 20, no. 9, p. 628, 2018

2018
[15]

Statistical distance and the geometry of quantum states,

S. L. Braunstein and C. M. Caves, “Statistical distance and the geometry of quantum states,” Physical Review Letters, vol. 72, no. 22, p. 3439, 1994

1994
[16]

Applications of the van Trees inequality: a Bayesian Cram´ er-Rao bound,

R. D. Gill and B. Y. Levit, “Applications of the van Trees inequality: a Bayesian Cram´ er-Rao bound,”Bernoulli, vol. 1, no. 1-2, pp. 59 – 79, 1995

1995
[17]

Optical atomic clocks,

A. D. Ludlow, M. M. Boyd, J. Ye, E. Peik, and P. O. Schmidt, “Optical atomic clocks,” Reviews of Modern Physics, vol. 87, no. 2, pp. 637–701, 2015

2015
[18]

Optical magnetometry,

D. Budker and M. Romalis, “Optical magnetometry,”Nature physics, vol. 3, no. 4, pp. 227–234, 2007

2007
[19]

Quantum-enhanced optical-phase tracking,

H. Yonezawa, D. Nakane, T. A. Wheatley, K. Iwasawa, S. Takeda, H. Arao, K. Ohki, K. Tsumura, D. W. Berry, T. C. Ralphet al., “Quantum-enhanced optical-phase tracking,” Science, vol. 337, no. 6101, pp. 1514–1517, 2012

2012
[20]

Sensitivity optimization for nv-diamond magnetometry,

J. F. Barry, J. M. Schloss, E. Bauch, M. J. Turner, C. A. Hart, L. M. Pham, and R. L. Walsworth, “Sensitivity optimization for nv-diamond magnetometry,”Reviews of Modern Physics, vol. 92, no. 1, p. 015004, 2020

2020
[21]

C. W. Helstrom,Quantum detection and estimation theory. Academic press, 1976. [Online]. Available: https://www.sciencedirect.com/bookseries/ mathematics-in-science-and-engineering/vol/123/suppl/C

1976
[22]

Hayashi,Quantum Information Theory

M. Hayashi,Quantum Information Theory. Springer, 2016

2016
[23]

Multiparameter estimation with two-qubit probes in noisy channels,

L. O. Conlon, P. K. Lam, and S. M. Assad, “Multiparameter estimation with two-qubit probes in noisy channels,”Entropy, vol. 25, no. 8, p. 1122, 2023

2023
[24]

A. S. Holevo,Probabilistic and statistical aspects of quantum theory. Edizioni della Normale, 2011

2011
[25]

Asymptotic performance of optimal state estimation in qubit system,

M. Hayashi and K. Matsumoto, “Asymptotic performance of optimal state estimation in qubit system,”Journal of Mathematical Physics, vol. 49, no. 10, 2008

2008
[26]

State estimation for large ensembles,

R. D. Gill and S. Massar, “State estimation for large ensembles,”Physical Review A, vol. 61, no. 4, p. 042312, 2000. 19

2000
[27]

Strong consistency and asymptotic efficiency for adaptive quantum estimation problems,

A. Fujiwara, “Strong consistency and asymptotic efficiency for adaptive quantum estimation problems,”Journal of Physics A: Mathematical and General, vol. 39, no. 40, p. 12489, 2006

2006
[28]

H. L. Van Trees,Detection, estimation, and modulation theory, part I: detection, estimation, and linear modulation theory. John Wiley & Sons, 2004

2004
[29]

Application of quantum estimation theory to analog communication over quantum channels,

S. Personick, “Application of quantum estimation theory to analog communication over quantum channels,”IEEE Transactions on Information Theory, vol. 17, no. 3, pp. 240–246, 1971

1971
[30]

Quantum metrology in the presence of limited data,

J. Rubio and J. Dunningham, “Quantum metrology in the presence of limited data,”New Journal of Physics, vol. 21, no. 4, p. 043037, 2019

2019
[31]

Multi-parameter estimation beyond quantum fisher information,

R. Demkowicz-Dobrza´ nski, W. G´ orecki, and M. Gut ¸˘ a, “Multi-parameter estimation beyond quantum fisher information,”Journal of Physics A: Mathematical and Theoretical, vol. 53, no. 36, p. 363001, 2020

2020
[32]

Bayesian logarithmic derivative type lower bounds for quantum estimation,

J. Zhang and J. Suzuki, “Bayesian logarithmic derivative type lower bounds for quantum estimation,”arXiv preprint arXiv:2405.10525, 2024

2024 arXiv
[33]

The hybrid cramer-rao lower bound - from practice to theory,

H. Messer, “The hybrid cramer-rao lower bound - from practice to theory,” inFourth IEEE Workshop on Sensor Array and Multichannel Processing, 2006., 2006, pp. 304–307

2006
[34]

Performance bounds for parameter estimation under misspecified models: Fundamental findings and applications,

S. Fortunati, F. Gini, M. S. Greco, and C. D. Richmond, “Performance bounds for parameter estimation under misspecified models: Fundamental findings and applications,”IEEE Signal Processing Magazine, vol. 34, no. 6, pp. 142–157, 2017

2017
[35]

Quantum detection and estimation theory,

C. W. Helstrom, “Quantum detection and estimation theory,”Journal of Statistical Physics, vol. 1, pp. 231–252, 1969

1969
[36]

On optimal quantum receivers for digital signal detection,

H. Yuen, R. Kennedy, and M. Lax, “On optimal quantum receivers for digital signal detection,”Proceedings of the IEEE, vol. 58, no. 10, pp. 1770–1773, 1970

1970
[37]

Estimation of a structural parameter in the presence of a large number of nuisance parameters,

M. Kumon and S.-I. Amari, “Estimation of a structural parameter in the presence of a large number of nuisance parameters,”Biometrika, vol. 71, no. 3, pp. 445–459, 1984. [Online]. Available: http://www.jstor.org/stable/2336554

1984
[38]

Estimation in the Presence of Infinitely many Nuisance Parameters–Geometry of Estimating Functions,

S.-I. Amari and M. Kumon, “Estimation in the Presence of Infinitely many Nuisance Parameters–Geometry of Estimating Functions,”The Annals of Statistics, vol. 16, no. 3, pp. 1044 – 1068, 1988. [Online]. Available: https://doi.org/10.1214/aos/1176350947

1988
[39]

Information geometrical characterization of quantum statistical models in quantum estimation theory,

J. Suzuki, “Information geometrical characterization of quantum statistical models in quantum estimation theory,”Entropy, vol. 21, no. 7, p. 703, 2019

2019
[40]

Bhatia,Positive definite matrices

R. Bhatia,Positive definite matrices. Princeton university press, 2009. 20

2009

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.

[1] [1]

Quantum sensing,

C. L. Degen, F. Reinhard, and P. Cappellaro, “Quantum sensing,”Reviews of Modern Physics, vol. 89, no. 3, p. 035002, 2017

2017

[2] [2]

Advances in quantum metrology,

V. Giovannetti, S. Lloyd, and L. Maccone, “Advances in quantum metrology,”Nature photonics, vol. 5, no. 4, p. 222, 2011

2011

[3] [3]

Quantum metrology with nonclassical states of atomic ensembles,

L. Pezze, A. Smerzi, M. K. Oberthaler, R. Schmied, and P. Treutlein, “Quantum metrology with nonclassical states of atomic ensembles,”Reviews of Modern Physics, vol. 90, no. 3, p. 035005, 2018

2018

[4] [4]

Quantum estimation for quantum technology,

M. G. A. Paris, “Quantum estimation for quantum technology,”International Journal of Quantum Information, vol. 7, pp. 125–137, 2009

2009

[5] [5]

The elusive heisenberg limit in quantum-enhanced metrology,

R. Demkowicz-Dobrza´ nski, J. Ko lody´ nski, and M. Gut ¸˘ a, “The elusive heisenberg limit in quantum-enhanced metrology,”Nature communications, vol. 3, no. 1, p. 1063, 2012

2012

[6] [6]

Critical quantum metrology in the non-linear quantum rabi model,

Z.-J. Ying, S. Felicetti, G. Liu, and D. Braak, “Critical quantum metrology in the non-linear quantum rabi model,”Entropy, vol. 24, no. 8, p. 1015, 2022

2022

[7] [7]

Nuisance parameter problem in quantum estimation theory: Tradeoff relation and qubit examples,

J. Suzuki, “Nuisance parameter problem in quantum estimation theory: Tradeoff relation and qubit examples,”Journal of Physics A: Mathematical and Theoretical, vol. 53, no. 26, p. 264001, 2020

2020

[8] [8]

Quantum state estimation with nuisance parameters,

J. Suzuki, Y. Yang, and M. Hayashi, “Quantum state estimation with nuisance parameters,” Journal of Physics A: Mathematical and Theoretical, vol. 53, no. 45, p. 453001, 2020. 18

2020