Pith. sign in

REVIEW 3 major objections 5 minor 27 references

Continuous-Variable Quantum Key Distribution with a Real Local Oscillator and without Auxiliary Signals

T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read The paper reports the first experimental demonstration of continuous-variable quantum key distribution with a real local oscillator and no auxiliary pilot signals, achieving an asymptotic key rate of 9.2 Mbit/s over 26 km of fiber.

desk verdict First pilot-free real-LO CV-QKD experiment is impressive, but the headline key rate leans on an unproven security assumption. read the letter →

arxiv 1908.03625 v1 pith:WC5ESOGM submitted 2019-08-02 quant-ph cs.CRcs.LGeess.SPphysics.optics

Sebastian Kleis , Max Rueckmann , Christian G. Schaeffer This is my paper · ORCID
classification quant-phcs.CRcs.LGeess.SPphysics.optics
keywords continuous-variablequantumkeydistributionreallocaloscillatorpilot-freephaserecoveryparticlesmootherMCMCparameteroptimizationdiscretemodulationcarrierestimationcoherentopticalcommunications
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

Continuous-variable quantum key distribution (CV-QKD) reads keys from faint laser pulses, so the receiver must track the laser phase at signal-to-noise ratios far below 0 dB. The usual fixes are a remote local oscillator or dedicated pilot tones; this paper removes both. It reports the first experimental demonstration of CV-QKD with a real free-running local oscillator and no auxiliary signals, using a machine-learning phase tracker. With the M=4 discrete-modulation protocol and only 5% of symbols revealed, the system demodulates down to a quantum-channel SNR of -19.1 dB and supports an asymptotic key rate of 9.2 Mbit/s over 26 km of fiber. If correct, this makes QKD hardware almost identical to a standard coherent optical link.

What carries the argument

The load-bearing element is the Bayesian particle smoother used for carrier phase estimation. It treats the laser phase as a random walk with a drifting frequency offset and, crucially, uses different measurement likelihoods for revealed and unrevealed symbols: when Alice has publicly revealed symbol a_k, the likelihood is a Gaussian centered on that known symbol; otherwise it is a mixture over the M possible symbols. A bootstrap particle filter with 200 particles plus a backward-simulation smoother performs the tracking, and MCMC optimization fixes the state-space noise variances. The same state-space model is reused in an extended Kalman filter for timing recovery. This machinery lets the receiver track the phase at SNR values where decision-directed methods fail.

What would settle it

Re-evaluate the secret key rate using a security analysis in which Bob's phase estimate is explicitly a function of the revealed-symbol sequence, with the same measured excess noise (signal-to-distortion ratio about 31.82 at SNRb = 7.1 dB) and receiver parameters; if the rate at 26 km falls below 9.2 Mbit/s, the pilot-free claim is falsified. A simpler marker: reproduce the -19.1 dB sensitivity by running the same particle-smoother DSP on independently recorded 17 GBd data.

Watch

Extended reading notes

Core claim

The paper's central claim is that pilot-free real-local-oscillator CV-QKD is experimentally feasible. Bob's free-running laser serves as the local oscillator, and carrier phase estimation is performed by a Bayesian particle smoother whose measurement model switches depending on whether each received symbol was publicly revealed by Alice. Because a fraction of symbols must be revealed in any CV-QKD protocol anyway, these revealed symbols double as phase references at no extra bandwidth cost. With M=4 modulation at 17 GBd, the measured system reaches an SNR of -19.1 dB and, using the measured excess noise and standard discrete-modulation security analysis, gives an asymptotic key rate of 5.7e-4 bit/sym, i.e., 9.2 Mbit/s over 26 km. Simulations limited to laser phase noise indicate the same receiver concept could extend to about 72 km for M=4.

Load-bearing premise

The key-rate claim assumes that the standard discrete-modulation security proofs remain valid when Bob's phase estimate is itself shaped by the publicly revealed symbols, a dependence the proofs do not explicitly model.

Editorial extensions

If this is right

  • Pilot tones are not indispensable for real-local-oscillator CV-QKD; a receiver that uses only the quantum signal and the already-revealed symbols can operate at deeply negative SNR.
  • Because the transmitter and receiver are almost identical to a classical coherent fiber link, existing commercial components and digital signal processing can be reused for QKD.
  • At the demonstrated parameters (M=4, pr=0.05, 26 km), the asymptotic key rate is 9.2 Mbit/s; simulations suggest the phase-noise-limited ceiling is around 72 km for the same modulation.
  • Increasing the revelation probability pr reduces excess noise and can tolerate stronger laser phase noise, trading a small fraction of public symbols for longer distance or cheaper lasers.

Reading between the lines

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

  • A direct security analysis that models Bob's phase estimator as a function of the revealed symbols would close the gap between the experiment and the proof; until then, the claimed key rate rests on an extrapolation of existing discrete-modulation proofs.
  • The same particle-smoother and MCMC phase tracking could be applied to other ultra-low-SNR coherent receivers, such as free-space optical satellite downlinks, where pilot tones are costly.
  • A real-time implementation of the particle smoother would be the next test; the paper's digital signal processing is offline, so latency and throughput at line rate remain open.
  • If the security proof is updated, the pilot-free design could roughly double the spectral efficiency of pilot-based real-local-oscillator systems, since no pilot bandwidth is reserved.
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.

Referee Report

3 major / 5 minor

Summary. This manuscript reports an experimental continuous-variable QKD setup with a real (local) oscillator and no pilot/auxiliary tones, using M-ary phase modulation and a Bayesian particle smoother for carrier phase recovery. The authors report successful demodulation of an M=4 discrete-modulation signal down to a channel SNR of -19.1 dB, and compute asymptotic secret-key rates of 9.2 Mbit/s over 26 km of fiber based on measured excess noise and receiver parameters. They also present simulations of a phase-noise-limited version of the system that reach longer distances, and they compare the experimental hard-decision mutual information with theoretical AWGN values.

Significance. The experimental result is significant: if a CV-QKD system with a free-running real LO can operate without pilot tones at SNR below -20 dB, it removes a major practical obstacle and brings CV-QKD closer to a standard coherent receiver. The evidence for the demodulation claim is credible: mutual information measurements are compared to theory, the excess-noise measurements show the expected scaling, and the authors are transparent about the limitations of their security analysis. The weakest step is not the demodulation physics but the transition from measured mutual information to a secret-key rate. The paper itself notes that the security proofs assume a linear Gaussian channel, and the receiver DSP is nonlinear and depends on the publicly revealed symbols. The headline key-rate number should therefore be treated as a conditional projection until this security gap is addressed.

major comments (3)
  1. [Discussion; Methods, Particle Smoother] The headline key rate is computed by applying the discrete-modulation security proofs of Refs. [1,23] to the measured parameters. The paper itself states in the Discussion that these proofs 'assume a linear Gaussian channel and are therefore not fully general,' and that assumption is exactly the problem here. The receiver output is produced by a particle smoother whose measurement model changes with the publicly revealed symbols r_k (Methods, Eqs. (10) and (11)) and which performs backward-simulation smoothing; the final symbols are thus a nonlinear, non-memoryless function of the raw measurements and of a time-correlated phase process. The manuscript does not show that the effective conditional distribution of Bob's key symbols after this post-processing is a linear Gaussian channel with the estimated transmission and excess noise, nor that Eve's optimal information is bounded by the Gaussian formulas used in [1,23]. Because the 9.2 Mbit/s at 26 km figure is an asymptotic secret-key rate, this is a load-bearing gap. I recommend either extending the security analysis to the actual DSP, or clearly labeling the key-rate number as a heuristic projection under an unproven channel-model assumption.
  2. [Discussion; Figure 5] The key-rate curves in Fig. 5 are computed from 'polynomial fits for the excess noise in the case of pr = 0.05' plotted in Figs. 3 and 4, but the measured excess-noise points are shown without error bars and the receiver parameters eta ≈ 0.232 and epsilon_el ≈ 0.70 enter through Eq. (4) without uncertainties. The excess noise is the parameter that most directly controls Eve's information, so the quantitative claim of 9.2 Mbit/s over 26 km has no stated statistical or systematic uncertainty. At a minimum, the authors should report the spread of the block-wise excess-noise estimates and propagate the calibration uncertainties (or provide a sensitivity analysis) through the key-rate calculation.
  3. [Results; Methods, Simulation Details; Figure 2] The simulations are described as showing the phase-noise-limited performance and are used to project distances up to 72 km, but they are not an independent test: the phase-noise sequence is recorded with the experimental setup, and the state-space parameters sigma_Omega^2 and sigma_phi^2 are fitted to a high-SNR (11.5 dB), pr = 1 measurement with the same lasers. The simulation curves therefore represent a best-case extrapolation, and the distinction between measured and simulated key rates should be maintained in the conclusions. In particular, the sentence in the conclusion that 'the achievable key rate over a distance of 26 km is 5.7 × 10^-4 bit/sym corresponding to 9.2 Mbit/s' should be qualified as an asymptotic, simulation- and fit-based projection rather than a directly measured rate.
minor comments (5)
  1. [Methods, Estimation of Mutual Information] In Eq. (14), the exponent in the last factor appears as exp(-SNR sin^2(...)); the subscript b in SNR_b is missing. Please correct this typo.
  2. [Figure 4 caption] The caption says 'The small and solid markers indicate negative values' but the marker shapes and colors are not defined in the legend; please clarify which marker style corresponds to which quantity.
  3. [Abstract and Conclusion] The abstract and conclusion state the 9.2 Mbit/s figure without the qualifier 'asymptotic' that is introduced in the Discussion. Please use the qualifier consistently so that the asymptotic nature of the key-rate calculation is clear to the reader.
  4. [Methods, Experimental Details] The text contains a character-encoding artifact ('AliceâĂŹs') and the Discussion contains the misspelling 'sensitvity'. These should be corrected.
  5. [Abstract] The abstract credits 'Machine Learning methods'; the paper actually uses Bayesian filtering and smoothing with MCMC parameter optimization. Please describe the methods more precisely to avoid overclaiming the novelty of the algorithmic approach.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the key-rate and distance projections are explicit calculations from measured excess noise and security proofs, and the security-proof applicability limitation is a validity gap, not a circular reduction.

full rationale

The central experimental claim—successful demodulation at SNR_b = −19.1 dB with pr = 0.05 and no pilot tones—is a direct measurement, not derived from the key-rate formula. The asymptotic key rates are presented as calculations, not independent predictions: the paper states, 'For this, we used the polynomial fits for the excess noise in the case of pr = 0.05 that are plotted in the figures 3 and 4.' Feeding fitted excess noise into an external security proof is a standard system-evaluation procedure, and the key-rate formula is not defined in terms of the measured key rate, so there is no reduction by construction. The simulation projections use a phase-noise sequence 'taken from a measurement with the experimental setup,' which is an independent input to the simulation rather than the output being predicted. Self-citations ([2], [22], [25]) are present, and Eq. (5) for excess noise is attributed to the authors' prior work, but the relation is a standard trigonometric expression and is not load-bearing in a way that forces the central claim. The most serious concern is correctly flagged by the authors themselves: the security proofs [1,23] 'assume a linear Gaussian channel and are therefore not fully general,' while the particle-smoother receiver uses publicly revealed symbols and nonlinear post-processing. That is an external-validity or correctness gap in the security argument, not circularity, because the paper does not claim the Gaussian channel model is derived from the protocol; it explicitly acknowledges the mismatch. Overall, the derivation chain is self-contained with respect to circularity, with only minor reliance on the authors' earlier excess-noise relation.

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

The central claim rests on a small set of fitted parameters (phase-noise variances, excess-noise polynomial fits) and on security and channel-model assumptions that are standard in CV-QKD but not re-validated for this pilotless architecture. No new physical entities are introduced.

free parameters (5)
  • sigma_Omega^2 (phase frequency drift variance) = 1.66e-16 rad^2
    Fitted via MCMC with an EKF on a high-SNR (11.5 dB) received signal; used in the state-space model for phase estimation and in simulations.
  • sigma_phi^2 (laser phase noise variance) = 6.36e-9 rad^2
    Same fitting procedure as sigma_Omega^2; used in the particle smoother and simulations.
  • Polynomial fits for excess noise xi_b (pr=0.05) = not given in text
    Polynomial fits to the experimental excess noise vs. SNR curves are used to compute the asymptotic key rates in figure 5; the coefficients are not listed.
  • Receiver efficiency eta = 0.232
    Measured calibration parameter used in the key-rate calculation and in the relation between received power and SNR.
  • Electrical-to-shot-noise ratio epsilon_el = 0.70
    Measured calibration parameter; affects the received power penalty and the excess noise simulation.
assumptions (5)
  • domain assumption Security proofs for discrete-modulation CV-QKD (Leverrier-Grangier [1, 23]) apply, which assume a linear Gaussian channel and trusted excess noise.
    The paper relies on these proofs for the key-rate calculation and notes the linear-Gaussian limitation, citing [24] as a recent relaxation.
  • ad hoc to paper The laser phase and frequency drift follow a random-walk model (Eqs. 12-13).
    The state-space model used by the particle smoother and EKF is assumed; its parameters are fitted to a measured signal.
  • domain assumption The AWGN channel model with known noise power (Eq. 9) holds for the received symbols.
    The particle smoother measurement model assumes white Gaussian noise with mean power PTN; any non-Gaussian or correlated noise is not modeled.
  • domain assumption Using publicly revealed symbols for phase estimation does not open a side channel that breaks the security proof.
    The paper does not analyze this; the key-rate calculation implicitly assumes the phase estimation is a trusted internal operation.
  • domain assumption The estimated excess noise in Bob's receiver bounds Eve's information (standard CV-QKD trusted-device assumption).
    Key-rate formulas use measured excess noise as if it fully accounts for eavesdropping information.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Continuous-Variable Quantum Key Distribution with a Real Local Oscillator and without Auxiliary Signals." pith.science (2026). https://pith.science/paper/WC5ESOGM

@misc{pith2026190803625,
  author       = {Pith},
  title        = {Pith review of: Continuous-Variable Quantum Key Distribution with a Real Local Oscillator and without Auxiliary Signals},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WC5ESOGM}},
  note         = {Machine review of arXiv:1908.03625}
}
read the original abstract

Continuous-variable quantum key distribution (CV-QKD) is realized with coherent detection and is therefore very suitable for a cost-efficient implementation. The major challenge in CV-QKD is mitigation of laser phase noise at a signal to noise ratio of much less than 0 dB. So far, this has been achieved with a remote local oscillator or with auxiliary signals. For the first time, we experimentally demonstrate that CV-QKD can be performed with a real local oscillator and without auxiliary signals which is achieved by applying Machine Learning methods. It is shown that, with the most established discrete modulation protocol, the experimental system works down to a quantum channel signal to noise ratio of -19.1 dB. The performance of the experimental system allows CV-QKD at a key rate of 9.2 Mbit/s over a fiber distance of 26 km. After remote local oscillator and auxiliary signal aided CV-QKD, this could mark a starting point for a third generation of CV-QKD systems that are even more attractive for a wide implementation because they are almost identical to standard coherent systems.

Figures

Figures reproduced from arXiv: 1908.03625 by the authors.

Figure 1
Figure 1. Experimental setup and digital signal processing (DSP) routine for the CV-QKD system [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Block diagram for the simulations. The measured phase was obtained using the experi [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Resulting system performance for M = 4 in terms of the mutual information between Alice and Bob IAB and the excess noise in Bob’s received symbols ξ 0 b . Experimental results are shown on the left hand side (a). Simulation results are shown on the right hand side (b). of the simulation model is shown in figure 2. As only additive noise and laser phase noise are included in the simulation model, it does indeed corre… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Resulting system performance for M = 2 in terms of the mutual information between Alice and Bob IAB and the excess noise in Bob’s received symbols ξ 0 b . Experimental results are shown on the left hand side (a). Simulation results are shown on the right hand side (b).…
Figure 5
Figure 5. Figure 5: Achievable key rates for pr = 0.05, taking into account the polynomial fits for the excess noise in figure 3 as well as the experimental parameters ηˆ, εˆel and SNRb,min. A reconciliation efficiency of 95 % and a fiber loss of 0.2 dB/km was assumed. The launch power wa…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 24 canonical work pages

  1. [1]

    Bob uses a balanced receiver to perform heterodyne detection at an intermediate frequency ofδνAB≈ 10 GHz

    The quantum signal is discrete phase-modulated in baseband with a modulation order M = 2 or M = 4 at a symbol rate of17 GBd. Bob uses a balanced receiver to perform heterodyne detection at an intermediate frequency ofδνAB≈ 10 GHz. In the DSP routine, Bob first down-converts the signal to baseband and applies the matched filterhQ. Then, the signal is down-sa...

    work page

  2. [2]

    Leverrier and P

    A. Leverrier and P. Grangier, Continuous-variable quantum-key-distribution protocols with a non-Gaussian modulation, Physical Review A83, 042312 (2011)

    work page 2011

  3. [3]

    Kleis and C

    S. Kleis and C. G. Schaeffer, Influence of the SNR of Pilot Tones on the Carrier Phase Estima- tion in Coherent Quantum Receivers, in2017 European Conference on Optical Communication (ECOC) (IEEE, Gothenburg, 2017) pp. 1–3

    work page 2017

  4. [4]

    Grosshans, G

    F. Grosshans, G. Van Assche, J. Wenger, R. Brouri, N. J. Cerf, and P. Grangier, Quantum key distribution using gaussian-modulated coherent states, Nature421, 238 (2003)

    2003

  5. [5]

    Lodewyck, T

    J. Lodewyck, T. Debuisschert, R. Tualle-Brouri, and P. Grangier, Controlling excess noise in fiber-optics continuous-variable quantum key distribution, Physical Review A72, 050303 (2005)

    work page 2005

  6. [6]

    B. Qi, L. Huang, and H. Lo, Experimental study on Gaussian-modulated coherent states quantum key distribution over standard telecom fiber, Physical Review A76, 052323 (2007)

    work page 2007

  7. [7]

    P.Jouguet, S.Kunz-Jacques, A.Leverrier, P.Grangier,andE.Diamanti,Experimentaldemon- stration of long-distance continuous-variable quantum key distribution, Nature Photonics7, 378 (2013)

    work page 2013

  8. [8]

    Jouguet, S

    P. Jouguet, S. Kunz-Jacques, and E. Diamanti, Preventing calibration attacks on the local oscillator in continuous-variable quantum key distribution, Physical Review A 87, 062313 (2013)

    work page 2013

Show all 27 references
  1. [9]

    Kleis, R

    S. Kleis, R. Herschel, and C. G. Schaeffer, Simple and Efficient Detection Scheme for Contin- uous Variable Quantum Key Distribution with m-ary Phase-Shift-Keying, inCLEO: Science 16 and Innovations (Optical Society of America, 2015) pp. SW3M–7

    2015

  2. [10]

    Huang, P

    D. Huang, P. Huang, D. Lin, C. Wang, and G. Zeng, High-speed continuous-variable quantum key distribution without sending a local oscillator, Optics Letters40, 3695 (2015)

    2015

  3. [11]

    B. Qi, P. Lougovski, R. Pooser, W. Grice, and M. Bobrek, Generating the Local Oscillator “Locally” in Continuous-Variable Quantum Key Distribution Based on Coherent Detection, Physical Review X5, 10.1103/PhysRevX.5.041009 (2015)

    2015 doi

  4. [12]

    D. B. Soh, C. Brif, P. J. Coles, N. Lütkenhaus, R. M. Camacho, J. Urayama, and M. Sarovar, Self-Referenced Continuous-Variable Quantum Key Distribution Protocol, Physical Review X 5, 10.1103/PhysRevX.5.041010 (2015)

    2015 doi

  5. [13]

    Schrenk and H

    B. Schrenk and H. Hübel, Pilot-Assisted Local Oscillator Synchronisation for CV-QKD, in6th International Conference on Quantum Cryptography (QCrypt)(2016)

    2016

  6. [14]

    Marie and R

    A. Marie and R. Alléaume, Self-coherent phase reference sharing for continuous-variable quan- tum key distribution, Physical Review A95, 012316 (2017)

    2017

  7. [15]

    Schrenk, F

    B. Schrenk, F. Laudenbach, C.-H. F. Fung, C. Pacher, A. Poppe, R. Lieger, D. Hillerkuss, E. Querasser, G. Humer, M. Hentschel, M. Peev, and H. Hübel, High-Rate Continuous- Variables Quantum Key Distribution with Piloted-Disciplined Local Oscillator, inECOC 2017; 43rd European ...

    2017

  8. [16]

    L. C. Comandar, H. H. Brunner, F. Karinou, F. Fung, S. Bettelli, D. Hillerkuss, S. Mikroulis, D. Wang, M. Kuschnerov, C. Xie, A. Poppe, and M. Peev, A flexible continuous-variable QKD system using off-the-shelf components, inQuantum Information Science and Technology III, edited...

    2017

  9. [17]

    Laudenbach, B

    F. Laudenbach, B. Schrenk, C. Pacher, M. Hentschel, C.-H. F. Fung, F. Karinou, A. Poppe, M. Peev, and H. Hübel, Pilot-assisted intradyne reception for high-speed continuous-variable quantum key distribution with true local oscillator, arXiv e-prints , 1712.10242 (2017)

    2017 arXiv

  10. [18]

    T. Wang, P. Huang, Y. Zhou, W. Liu, H. Ma, S. Wang, and G. Zeng, High key rate continuous- variable quantum key distribution with a real local oscillator, Optics Express26, 2794 (2018)

    2018

  11. [19]

    Fossier, E

    S. Fossier, E. Diamanti, T. Debuisschert, R. Tualle-Brouri, and P. Grangier, Improvement of continuous-variable quantum key distribution systems by using optical preamplifiers, Journal of Physics B: Atomic, Molecular and Optical Physics42, 114014 (2009)

    2009

  12. [20]

    Grosshans, N

    F. Grosshans, N. J. Cerf, J. Wenger, R. Tualle-Brouri, and P. Grangier, Virtual Entanglement and Reconciliation Protocols for Quantum Cryptography with Continuous Variables, Quantum 17 information & computation3, 535 (2003)

    2003

  13. [21]

    Oerder and H

    M. Oerder and H. Meyr, Digital Filter and Square Timing Recovery, IEEE Transactions on Communications 36, 605 (1988)

    1988

  14. [22]

    Särkkä,Bayesian filtering and smoothing, Vol

    S. Särkkä,Bayesian filtering and smoothing, Vol. 3 (Cambridge University Press, 2013)

    2013

  15. [23]

    Kleis and C

    S. Kleis and C. G. Schaeffer, Improving the Secret Key Rate of Coherent Quantum Key Dis- tribution with Bayesian Inference, Journal of Lightwave Technology37, 722 (2019)

    2019

  16. [24]

    Leverrier and P

    A. Leverrier and P. Grangier, Continuous-variable quantum key distribution protocols with a discrete modulation, arXiv e-prints , 1002.4083 (2010)

    2010 arXiv

  17. [25]

    Ghorai, P

    S. Ghorai, P. Grangier, E. Diamanti, and A. Leverrier, Asymptotic security of continuous- variable quantum key distribution with a discrete modulation, Physical Review X9, 021059 (2019)

    2019

  18. [26]

    Piels, M

    M. Piels, M. I. Olmedo, W. Xue, X. Pang, C. Schaffer, R. Schatz, G. Jacobsen, I. T. Monroy, J. Mork, S. Popov, and D. Zibar, Laser Rate Equation-Based Filtering for Carrier Recovery in Characterization and Communication, Journal of Lightwave Technology33, 3271 (2015)

    2015

  19. [27]

    J. C. Lagarias, J. A. Reeds, M. H. Wright, and P. E. Wright, Convergence Properties of the Nelder–Mead Simplex Method in Low Dimensions, SIAM Journal on Optimization9, 112 (1998). 18

    1998

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.