Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

Parameter-Shift Rules for Gradients in Boson Sampling Experiments

T0 review · 3 major / 5 minor · reviewed 2026-08-02 · deepseek-v4-flash

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

Pith's one-line read In Fock boson sampling under loss, transition probabilities are finite Fourier series in each phase, enabling exact gradients via n-th order parameter-shift rules; general GBS admits no such rule.

desk verdict The Fock PSR result is real and the GBS boundary is new, but the paper overclaims on hardware exactness and the GBS no-go; worth refereeing with revision requests on scope and claims. read the letter →

arxiv 2607.15160 v1 pith:6A6BIHVW submitted 2026-07-16 quant-ph

Marius Trudeau , Pierre-Emmanuel Emeriau , Nicolás Quesada This is my paper · ORCID
classification quant-ph
keywords parameter-shiftrulesFockbosonsamplingGaussianphotonlossgradientestimationpermanenthafnianvariationalquantumalgorithms
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

Fock boson sampling on a lossy interferometer retains a hidden structure: every transition probability is a finite Fourier series in any one tunable phase, with frequencies bounded by the number of input photons. This paper shows that structure makes the exact gradient computable by an n-th order parameter-shift rule — evaluating the circuit at shifted phases and linearly combining the results — even when the device is lossy and noisy. For Gaussian boson sampling the same strategy generally fails, because squeezed states have unbounded photon-number support and loss mixes in infinitely many frequency terms; an exact finite-order rule survives only when loss can be pushed before the interferometer. The authors verify the rule on simulated and real hardware, where it is markedly more stable than finite differences. This matters for variational photonic algorithms, which need reliable gradients to optimize circuits on the device itself.

What carries the argument

The load-bearing identity is the permanent representation of the lossy Fock transition probability, (1/(i! j!)) Perm[B^{i⊕j}], together with the fact that each entry of the transmission matrix is a single-harmonic function of any given phase, T_mn = a e^{iθ}+b. From this, the entries of the permanent matrix B take the form a' e^{iθ}+b' e^{−iθ}+c', and because the permanent is a polynomial of degree equal to the total photon number n, the whole probability becomes a finite Fourier series with frequencies confined to [−n,n]. The parameter-shift rule exploits this structure by differentiating the series and solving a small linear system (Eqs. 38–43) to get exact derivative coefficients from shi

What would settle it

Measure the complex transmission matrix T(θ) of a real interferometer as a function of one phase (coherent-state input and heterodyne detection). If any entry deviates from a e^{iθ}+b — for example, amplitude varies with θ — the n-th order shift rule will show a systematic bias that scales with that coupling. Alternatively, for the GBS claim, measure a squeezed-state output probability versus θ under general loss and fit a finite Fourier series; if a finite order fits exactly, the claimed impossibility is wrong.

Watch

Extended reading notes

Core claim

For a lossy interferometer with transmission matrix T, when each tunable phase θ enters entries as T_mn = a e^{iθ}+b with a,b independent of θ, the Fock-state transition probability from input i to output j equals (1/(i! j!)) Perm[B^{i⊕j}], where B is constructed from T and E = I − T†T. Because a permanent of size n is a polynomial in its entries, the probability is a finite Fourier series Σ_{m=−n}^{n} k_m e^{imθ}, and its derivative can be reconstructed exactly from shifted circuit evaluations via an n-th order parameter-shift rule. The analogous Gaussian boson sampling probability is a hafnian expression divided by the square root of a determinant (Eqs. 90–94); under general loss these fac

Load-bearing premise

The rule is exact only if each tunable phase appears exactly once in the circuit and its amplitude transmission is independent of the phase value, so that T entries are a e^{iθ}+b; if real phase-shifters couple amplitude to phase, the finite Fourier bound and the n-th order rule break down.

Editorial extensions

If this is right

  • On any device where the phase-shifter model holds, the gradient of a Fock-state transition probability is computable exactly from 2n+1 (or n with symmetric shifts) circuit evaluations, independent of the number of modes or the loss level.
  • Variational photonic algorithms can replace finite differences with this shift rule, removing the step-size trade-off and gaining robustness to phase noise, as demonstrated numerically and on Quandela's QPU.
  • For Gaussian boson sampling, the result is a no-go: under general loss no exact finite-order rule exists; users must either engineer loss before the interferometer (where a rule of order 2|j| applies) or accept approximate alternative gradient estimators.
  • Threshold-detector probabilities inherit the same finite Fourier structure in the Fock case, since they are finite sums of Fock probabilities, so the rule extends to click patterns.
  • The order bound n = number of input photons gives a direct cost estimate: gradient evaluation scales linearly with input photon number, not with Hilbert-space dimension.

Reading between the lines

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

  • A testable extension suggested by the argument: the same finite-Fourier logic should apply to any photonic observable that is a degree-n polynomial in single-harmonic matrix entries, such as photon-number moments, multi-mode correlation functions, or averages of observables under Fock inputs.
  • The GBS no-go points to a practical truncation strategy: an approximate finite-order rule could be built by truncating the infinite Fock support of squeezed states, with an error that decays as the tail of the squeezed-state distribution; the paper does not explore this.
  • Because exactness rests on amplitude–phase decoupling, a hardware-level characterization of T(θ) along the lines of coherent-state tomography would provide a direct engineering criterion for when the method is applicable; the paper itself offers only an order-of-magnitude argument.
  • The permanent formula is derived without the Choi–Jamiołkowski trick, so the same derivation route might generalize to partial distinguishability or mixed input states, though the paper does not address those cases.
Share X Bluesky LinkedIn Reddit HN

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. The paper derives parameter-shift rules (PSRs) for gradient estimation in photonic boson sampling circuits with loss. For Fock-state inputs, it claims that each transition probability is a finite Fourier series in any tunable phase with frequencies bounded by the number of input photons n, so an n-th order PSR gives the exact derivative even under arbitrary loss. For Gaussian boson sampling (GBS), it claims that no finite-order PSR exists under general loss; a finite-order rule of order 2|j| exists only when the loss commutes through the interferometer (T = UΔ). The paper also presents numerical simulations and a hardware demonstration on Quandela's Belenos QPU comparing PSR with finite differences.

Significance. If the Fock-state result is correct, it is a useful and nontrivial generalization of PSRs: it removes the need for post-selection or unitary-only models when computing gradients on lossy photonic hardware. The derivation is largely self-contained, and the permanent/hafnian formulas are cross-checked against known results. The GBS no-go result, if made rigorous, would be an important limitation for variational methods with squeezed states. The hardware demonstration is a valuable practical check, although it does not by itself validate the exactness of the Fock PSR under the most general loss model.

major comments (3)
  1. [Sec. III.B, Eqs. (94)–(97)] The claim that no finite-order PSR exists for GBS under general loss is not rigorously established. The two supporting arguments — that det(Λ'_uu)^(-1/2) is not polynomial in the entries of T, and that Eq. (97) contains an infinite sum over photon numbers — are suggestive but not proofs. A non-polynomial dependence on T does not logically exclude cancellation that leaves a finite Fourier series in θ, and Eq. (97) is only an example (loss after the unitary), not a general impossibility. Since the abstract advertises this as a main result, please either supply a concrete counterexample (e.g., a simple two-mode circuit with generic loss where the probability has infinite Fourier support) or soften the claim to 'the standard PSR construction does not apply in general.'
  2. [Sec. II.B, Eq. (31)] The exactness of the Fock PSR rests on the assumption Φ_i^(j) = √η_{i,j} e^{iθ_{i,j}}, i.e., that each phase shifter's amplitude transmission is independent of the applied phase and that each θ appears exactly once. If η depends on θ (thermo-optic or electro-optic phase shifters can exhibit loss-phase coupling), then T entries no longer have the form a e^{iθ} + b, and the frequency bound in Eq. (66) can fail. The paper's physical justification is an order-of-magnitude statement about optical path length, not a device-level guarantee. Please either provide phase-shifter characterization data from Belenos/Perceval showing θ-independent transmission over the relevant range, or explicitly scope the headline claim to the model in Eq. (31).
  3. [Sec. III.A, Eq. (66)] The derivation of the frequency support [-n, n] is only sketched as 'three observations.' For a reader, the bound is not immediate because the permanent in Eq. (61) has size |i|+|j|, not |i|. I recommend explicitly stating the block-counting argument: in B^{i⊕j}, with m=|j|, every permutation must contain exactly m T-factors and m T†-factors and (|i|-m) E-factors; since T-factors and E-factors contribute at most +1 frequency and T†-factors contribute at most 0, the maximum positive frequency is m + (|i|-m) = |i|. This would make the main result easier to verify.
minor comments (5)
  1. [Eq. (94)] The prefactor is written as '1/⃗j!' but the state is |⃗k⟩; it should be '1/⃗k!'.
  2. [Fig. 3 and Fig. 4 captions] The vertical-axis labels are garbled ('| Pr( )|'); please fix the mathematical notation in the captions.
  3. [Sec. II.B] The phrase 'without loss of generality' before the assumption that phase shifters do not affect transmission is too strong; it is an assumption of the model, not a consequence of generality.
  4. [Sec. II.C after Eq. (38)] The ordering of indices in the linear system is unusual but acceptable; please add a sentence clarifying that the DFT choice μ_m = 2πm/(2n+1) yields an invertible system for all n, not just for the displayed ordering.
  5. [Sec. IV] The hardware experiment uses a two-mode HOM setup with uniform loss; this is a valid proof-of-principle but does not exercise the 'arbitrary loss' scenario of the main theorem. A sentence acknowledging this limitation would be appropriate.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: Fock and GBS gradient results are derived in-paper from explicit loss-model assumptions.

full rationale

The central derivation chain is self-contained. Equation (61) is obtained in-paper from the coherent-state generating function and hafnian formalism (Eqs. (46)-(60)), with the paper explicitly noting agreement with Ref. [19] only after deriving it. Equation (66) is a direct algebraic consequence of Eq. (64), which follows from the phase-shifter model Eq. (31) and the decomposition in Fig. 1. The PSR construction (Eqs. (32)-(43)) is re-derived rather than imported; the DFT shift-angle choice is attributed to Ref. [42], but exactness holds for any invertible shift matrix, so this author-overlapping citation is not load-bearing for the exactness claim. The GBS negative result is derived from the explicit expression Eq. (94) and illustrated by the infinite-support argument in Eq. (97), not taken from prior work. No parameters are fitted to data and then reported as predictions; the experimental section compares a deterministic PSR estimator with finite differences on hardware. The only load-bearing premise, Eq. (31)'s assumption of theta-independent transmission, is a stated modeling assumption supported by a physical plausibility argument; whether real thermo-optic or electro-optic phase shifters satisfy it is a correctness/robustness question, not a circularity. Therefore no circular step is present.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

No free parameters are fitted to data: PSR coefficients come from solving the deterministic linear system (Eq. 43), and the experimental section treats ε, δ, shot counts, and step size as fixed hyperparameters of the comparison, not fitted constants. Invented entities: none — no new particles, forces, or hidden sectors. The load-bearing assumptions are the vacuum-environment loss model (Eq. 26), the phase-independent transmission of phase-shifters (Eq. 31), and the decomposition structure of Fig. 1; all are standard in photonics but are modeled, not device-verified.

assumptions (7)
  • standard math Canonical bosonic formalism: commutation relations, Fock/coherent/squeezed states, unitary mode transformations (Eqs. 1-25)
    Standard quantum-optics background used throughout Sec. II; not contested.
  • domain assumption Loss model: arbitrary lossy interferometer = sub-unitary T embedded in a 2M-mode unitary W; environment in vacuum, traced out (Eqs. 26-29)
    Load-bearing for both the permanent (Eq. 61) and hafnian (Eq. 94) probability formulas; standard in the field but a modeling choice — no dark counts, no correlated/amplifying noise, no detector-dependent loss.
  • domain assumption Phase-shifter transmission is θ-independent: Φ = √η e^{iθ} (Eq. 31), with each θ entering T exactly once
    Load-bearing: gives T_{m,n} = a e^{iθ} + b (Eq. 64), the premise of the finite-Fourier argument. Justified in Sec. II.B by order-of-magnitude physics, not device verification.
  • domain assumption Universal interferometer decompositions factor into fixed sub-unitary blocks S_i and diagonal phase matrices (Fig. 1; Clements/Bell/DFT decompositions)
    Used to place all θ-dependence in diagonal phases; assumes loss in real devices distributes into the fixed blocks as drawn.
  • standard math Squeezed states have unbounded Fock support; per-photon Bernoulli loss after the unitary gives binomial mixtures (Eq. 97)
    Basis of the GBS no-PSR argument: infinitely many photon-number sectors contribute, giving infinitely many Fourier frequencies.
  • standard math Generating-function identities: coherent-state overlaps, hafnian/permanent replication (Eqs. 49-62)
    Underpins the derivation of Eqs. (61) and (94); standard combinatorics of Gaussian integrals.
  • standard math Woodbury matrix identity (Eq. 89)
    Used to simplify the GBS Σ matrix inversion; textbook result.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Parameter-Shift Rules for Gradients in Boson Sampling Experiments." pith.science (2026). https://pith.science/paper/6A6BIHVW

@misc{pith2026260715160,
  author       = {Pith},
  title        = {Pith review of: Parameter-Shift Rules for Gradients in Boson Sampling Experiments},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6A6BIHVW}},
  note         = {Machine review of arXiv:2607.15160}
}
abstract

Using a robust photon loss model for photonic quantum experiments, we derive $n-$th order parameter-shift rules for computing gradients of Fock boson sampling transition probabilities where $n$ photons are sent into a lossy interferometer. We also show that in general it is not possible to generalize this gradient recipe to the case of Gaussian boson sampling. Only in the specific case where the transmission matrix of the interferometer can be factorized as a diagonal loss matrix premultiplied by a pure unitary it is possible to obtain parameter-shift rules with a finite order given by twice the total number of photons detected. We demonstrate the efficacy of the proposed method by comparing its performance against finite differences on real hardware.

Figures

Figures reproduced from arXiv: 2607.15160 by the authors.

Figure 1
Figure 1. FIG. 1. A lossy interferometer decomposed as a sequence [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. Example of a sub-unitary matrix with diagonalizable [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Absolute derivative error as a function of the phase [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: FIG. 4. Derivative estimation error as a function of loss for FD with [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5. Cost function, corresponding to the coincidence prob [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. The trainability of photonic quantum circuits

    quant-ph 2026-07 conditional novelty 6.0 of 10

    Fixed-order photon-number polynomial observables make passive linear-optical variational circuits trainable with polynomially many samples, while output-probability and high-order observables are exponentially hard to train.

Reference graph

Works this paper leans on

63 extracted references · 8 canonical work pages · cited by 1 Pith paper

  1. [19]

    Riemannian optimization of photonic quantum circuits in phase and fock space,

    Y. Yao, F. Miatto, and N. Quesada, “Riemannian optimization of photonic quantum circuits in phase and fock space,”SciPost Physics, vol. 17, no. 3, p. 082, 2024. [Online]. Available: https://doi.org/10. 21468/SciPostPhys.17.3.082

    2024

  2. [42]

    Pho- tonic parameter-shift rule: Enabling gradient computa- tion for photonic quantum computers,

    A. Pappalardo, P.-E. Emeriau, G. de Felice, B. Ventura, H. Jaunin, R. Yeung, B. Coecke, and S. Mansfield, “Pho- tonic parameter-shift rule: Enabling gradient computa- tion for photonic quantum computers,”Phys. Rev. A, vol. 111, no. 3, p. 032429, 2025

    2025

  3. [1]

    Why i am optimistic about the silicon- photonic route to quantum computing,

    T. Rudolph, “Why i am optimistic about the silicon- photonic route to quantum computing,”APL Photonics, vol. 2, no. 3, p. 030901, Mar. 2017. [Online]. Available: https://doi.org/10.1063/1.4976737

    work page doi:10.1063/1.4976737 2017

  4. [2]

    Scalability of large-scale photonic integrated circuits,

    Y. Su, Y. He, X. Guo, W. Xie, X. Ji, H. Wang, X. Cai, L. Tong, and S. Yu, “Scalability of large-scale photonic integrated circuits,”ACS Photonics, vol. 10, no. 6, pp. 2020–2030, 2023

    2020

  5. [3]

    The potential and global outlook of integrated photonics for quantum technologies,

    E. Pelucchi, G. Fagas, I. Aharonovich, D. Englund, E. Figueroa, Q. Gong, H. H¨ ubel, J. Liu, C.-Y. Lu, N. Matsuda, J.-W. Pan, F. Schreck, F. Sciarrino, C. Silberhorn, J. Wang, and K. D. J¨ ons, “The potential and global outlook of integrated photonics for quantum technologies,”Nature Reviews Physics, vol. 4, pp. 194–208, 2022. [Online]. Available: https: ...

    doi:10.1038/s42254-021-00398-z 2022

  6. [4]

    Integrated photonic neural network with on-chip backpropagation training,

    F. Ashtiani, M. H. Idjadi, and K. Kim, “Integrated photonic neural network with on-chip backpropagation training,”Nature, 2026, published online: 18 March

    2026

  7. [5]

    A manufacturable platform for pho- tonic quantum computing,

    PsiQuantum Team, “A manufacturable platform for pho- tonic quantum computing,”Nature, vol. 641, pp. 876– 883, 2025, published online 26 February 2025

    2025

  8. [6]

    Scal- ing and networking a modular photonic quantum com- puter,

    H. Aghaee Rad, T. Ainsworth, R. N. Alexander, B. Al- tieri, M. F. Askarani, R. Baby, L. Banchi, B. Q. Bara- giola, J. E. Bourassa, R. S. Chadwick, I. Charania, H. Chen, M. J. Collins, P. Contu, N. D’Arcy, G. Dauphi- nais, R. De Prins, D. Deschenes, I. Di Luch, S. Duque, P. Edke, S. E. Fayer, S. Ferracin, H. Ferrettiet al., “Scal- ing and networking a modu...

    2025

Show all 63 references
  1. [7]

    A versatile single-photon-based quantum computing platform,

    N. Maring, A. Fyrillas, M. Pont, E. Ivanov, P. Stepanov, N. Margaria, W. Hease, A. Pishchagin, A. Lema ˆ ıtre, I. Sagnes, T. H. Au, S. Boissier, E. Bertasi, A. Baert, M. Valdivia, M. Billard, O. Acar, A. Brieussel, R. Mezher, S. C. Wein, A. Salavrakos, P. Sinnott, D. A. Fioret...

    2024

  2. [8]

    Quantum circuits with many photons on a programmable nanophotonic chip,

    J. M. Arrazola, V. Bergholm, K. Br¨ adler, T. R. Bromley, M. J. Collins, I. Dhand, A. Fumagalli, T. Gerrits, A. Goussev, L. G. Helt, J. Hundal, T. Isacsson, R. B. Israel, J. Izaac, S. Jahangiri, R. Janik, N. Killoran, S. P. Kumar, J. Lavoie, A. E. Lita, D. H. Mahler, M. Menott...

    2021

  3. [9]

    Integrated photonic quantum technologies,

    J. Wang, F. Sciarrino, A. Laing, and M. G. Thompson, “Integrated photonic quantum technologies,”Nature Photonics, vol. 14, pp. 273–284, 2020. [Online]. Available: https://doi.org/10.1038/s41566-019-0535-7

    2020 doi

  4. [10]

    Recent progress towards large-scale integrated photonic quantum computation,

    H. Zhu, T. Chen, H. Ma, Z. Zhao, J. Guo, Q. Liang, Y. Tu, T. He, W. Luo, B. Wang, Q. Song, C. Jin, F. Liu, Y. Yang, Y. Wang, K. C. Leong, D. Wang, and J. Yang, “Recent progress towards large-scale integrated photonic quantum computation,”npj Nanophotonics, vol. 3, no. 1, pp. 1...

    2026 doi

  5. [11]

    Advances in photonic quantum sensing,

    S. Pirandola, B. Bardhan, T. Gehring, C. Weedbrook, and S. Lloyd, “Advances in photonic quantum sensing,” Nature Photonics, vol. 12, no. 12, pp. 724–733, 2018

    2018

  6. [12]

    Secure quantum key distribution,

    H.-K. Lo, M. Curty, and K. Tamaki, “Secure quantum key distribution,”Nature Photonics, vol. 8, no. 8, pp. 595–604, 2014

    2014

  7. [13]

    Optimal design for universal multiport interferometers,

    W. R. Clements, P. C. Humphreys, B. J. Metcalf, W. S. Kolthammer, and I. A. Walmsley, “Optimal design for universal multiport interferometers,”Optica, vol. 3, no. 12, pp. 1460–1465, 2016. [Online]. Available: https://doi.org/10.1364/OPTICA.3.001460

    2016 doi

  8. [14]

    Etching properties of pin-pmn-pt and its integration in electro-optic phase modulators,

    S. Poveda-Hospital, N. Quesada, and Y.-A. Peter, “Etching properties of pin-pmn-pt and its integration in electro-optic phase modulators,”Optical Materials Express, vol. 15, no. 3, pp. 522–530, 2025. [Online]. Available: https://doi.org/10.1364/OME.542571

    2025 doi

  9. [15]

    Foundations of photonic quantum compu- tation,

    M. Bombardelli, G. Fleury, P. Lacomme, and B. Vulpescu, “Foundations of photonic quantum compu- tation,”arXiv preprint arXiv:2509.04266, 2025. [Online]. Available: https://doi.org/10.48550/arXiv.2509.04266

    2025 doi

  10. [16]

    Programmable photonic integrated meshes for modular generation of optical entanglement links,

    M. Dong, K. J. Palm, M. Zimmermann, A. Witte, D. Heim, H. Choi, D. Dominguez, G. Gilbert, G. Clark, A. J. Leenheer, M. Eichenfield, and D. Englund, “Programmable photonic integrated meshes for modular generation of optical entanglement links,”npj Quantum Information, vol. 9, n...

    2023 doi

  11. [17]

    Wideband two-port beam splitter of a binary fused- silica phase grating,

    B. Wang, C. Zhou, J. Feng, H. Ru, and J. Zheng, “Wideband two-port beam splitter of a binary fused- silica phase grating,”Appl. Optics, vol. 47, no. 22, pp. 4004–4008, 2008. [Online]. Available: https: //doi.org/10.1364/AO.47.004004

    2008 doi

  12. [18]

    Silicon optical modulators,

    G. T. Reed, G. Mashanovich, F. Y. Gardes, and D. J. Thomson, “Silicon optical modulators,”Nature Photonics, vol. 4, pp. 518–526, 2010. [Online]. Available: https://doi.org/10.1038/nphoton.2010.179

    2010 doi

  13. [20]

    Threshold detection statistics of bosonic states,

    J. F. F. Bulmer, S. Paesani, R. S. Chadwick, and N. Que- sada, “Threshold detection statistics of bosonic states,” Phys. Rev. A, vol. 106, no. 4, p. 043712, 2022

    2022

  14. [21]

    Simulating lossy and partially distinguish- able quantum optical circuits: Theory, algorithms, and applications to experiment validation and state prepara- tion,

    J. F. Bulmer, J. Mart ´ ınez-Cifuentes, B. A. Bell, and N. Quesada, “Simulating lossy and partially distinguish- able quantum optical circuits: Theory, algorithms, and applications to experiment validation and state prepara- tion,”Advanced Photonics, vol. 8, no. 1, p. 016010, 2026

    2026

  15. [22]

    Quan- tum sampling problems, bosonsampling and quantum supremacy,

    A. P. Lund, M. J. Bremner, and T. C. Ralph, “Quan- tum sampling problems, bosonsampling and quantum supremacy,”npj Quantum Information, vol. 3, p. 15, 2017

    2017

  16. [23]

    Toward scalable 13 boson sampling with photon loss,

    H. Wang, W. Li, X. Jiang, Y.-M. He, Y.-H. Li, X. Ding, M.-C. Chen, J. Qin, C.-Z. Peng, C. Schneider, M. Kamp, W.-J. Zhang, H. Li, L.-X. You, Z. Wang, J. P. Dowling, S. H¨ ofling, C.-Y. Lu, and J.-W. Pan, “Toward scalable 13 boson sampling with photon loss,”Physical Review Let-...

    2018

  17. [24]

    Variational quantum algo- rithms,

    M. Cerezo, A. Arrasmith, R. Babbush, S. C. Benjamin, S. Endo, K. Fujii, J. R. McClean, K. Mitarai, X. Yuan, L. Cincio, and P. J. Coles, “Variational quantum algo- rithms,”Nature Reviews Physics, vol. 3, no. 9, pp. 625– 644, 2021

    2021

  18. [25]

    Varia- tional quantum algorithms: Fundamental concepts, ap- plications and challenges,

    H. Qi, S. Xiao, Z. Liu, C. Gong, and A. Gani, “Varia- tional quantum algorithms: Fundamental concepts, ap- plications and challenges,”Quantum Information Pro- cessing, vol. 23, no. 7, p. 250, 2024

    2024

  19. [26]

    A variational eigenvalue solver on a photonic quantum pro- cessor,

    A. Peruzzo, J. McClean, P. Shadbolt, M.-H. Yung, X.-Q. Zhou, P. J. Love, A. Aspuru-Guzik, and J. L. O’Brien, “A variational eigenvalue solver on a photonic quantum pro- cessor,”Nature Communications, vol. 5, p. 4213, 2014

    2014

  20. [27]

    The variational quantum eigensolver: A review of methods and best practices,

    J. Tilly, H. Chen, S. Cao, D. Picozzi, K. Setia, Y. Li, E. Grant, L. Wossnig, I. Rungger, G. H. Booth, and J. Tennyson, “The variational quantum eigensolver: A review of methods and best practices,”Physics Reports, vol. 986, pp. 1–128, 2022

    2022

  21. [28]

    A quantum approximate optimization algorithm,

    E. Farhi, J. Goldstone, and S. Gutmann, “A quantum approximate optimization algorithm,”arXiv preprint arXiv:1411.4028, 2014

    2014 arXiv

  22. [29]

    A review on quantum approximate optimization algorithm and its variants,

    K. Blekos, D. Brand, A. Ceschini, C.-H. Chou, R.-H. Li, K. Pandya, and A. Summer, “A review on quantum approximate optimization algorithm and its variants,” Physics Reports, vol. 1068, pp. 1–66, 2024

    2024

  23. [30]

    Circuit-centric quantum classifiers,

    M. Schuld, A. Bocharov, K. M. Svore, and N. Wiebe, “Circuit-centric quantum classifiers,”Phys. Rev. A, vol. 101, no. 3, p. 032308, 2020

    2020

  24. [31]

    Supervised learning with quantum-enhanced fea- ture spaces,

    V. Havl ´ ıˇ cek, A. D. C´ orcoles, K. Temme, A. W. Har- row, A. W. Kandala, J. M. Chow, and J. M. Gam- betta, “Supervised learning with quantum-enhanced fea- ture spaces,”Nature, vol. 567, pp. 209–212, 2019

    2019

  25. [32]

    Continuous-variable quan- tum neural networks,

    N. Killoran, T. R. Bromley, J. M. Arrazola, M. Schuld, N. Quesada, and S. Lloyd, “Continuous-variable quan- tum neural networks,”Phys. Rev. Research, vol. 1, no. 3, p. 033063, 2019

    2019

  26. [33]

    The quest for a quantum neural network,

    M. Schuld, I. Sinayskiy, and F. Petruccione, “The quest for a quantum neural network,”Quantum Information Processing, vol. 13, pp. 2567–2586, 2014

    2014

  27. [34]

    Machine learning with quantum computers,

    M. Schuld and F. Petruccione, “Machine learning with quantum computers,”Springer Nature, 2021

    2021

  28. [35]

    Quantum neural network for quantum neu- ral computing,

    M.-G. Zhou, Z.-P. Liu, H.-L. Yin, C.-L. Li, T.-K. Xu, and Z.-B. Chen, “Quantum neural network for quantum neu- ral computing,”Research, vol. 6, p. Article 0134, 2023

    2023

  29. [36]

    Recent developments and applications in quantum neural network: A re- view,

    S. K. Jeswal and S. Chakraverty, “Recent developments and applications in quantum neural network: A re- view,”Archives of Computational Methods in Engineer- ing, vol. 26, pp. 793–807, 2019

    2019

  30. [37]

    Quantum neural network,

    M. V. Altaisky, “Quantum neural network,”arXiv preprint arXiv:quant-ph/0107012, 2001

    2001 arXiv

  31. [38]

    Efficient variational quan- tum simulator incorporating active error minimization,

    Y. Li and S. C. Benjamin, “Efficient variational quan- tum simulator incorporating active error minimization,” Phys. Rev. X, vol. 7, no. 2, p. 021050, 2017

    2017

  32. [39]

    Theory of variational quantum simulation,

    X. Yuan, S. Endo, Q. Zhao, Y. Li, and S. C. Benjamin, “Theory of variational quantum simulation,”Quantum, vol. 3, p. 191, 2019

    2019

  33. [40]

    Optimized numerical gradient and hessian estimation for variational quantum algorithms,

    Y. S. Teo, “Optimized numerical gradient and hessian estimation for variational quantum algorithms,”Phys. Rev. A, vol. 107, no. 4, p. 042421, 2023

    2023

  34. [41]

    Qugstep: Refining step size selection in gradient estimation for variational quantum algorithms,

    S. Liang, L. Zhu, X. Li, and C. Yang, “Qugstep: Refining step size selection in gradient estimation for variational quantum algorithms,”arXiv preprint arXiv:2503.14366, 2025

    2025 arXiv

  35. [43]

    Exact gradients for linear optics with single photons,

    G. Facelli, D. D. Roberts, H. Wallner, A. Makarovskiy, Z. Holmes, and W. R. Clements, “Exact gradients for linear optics with single photons,”arXiv preprint arXiv:2409.16369, 2024. [Online]. Available: https: //arxiv.org/abs/2409.16369

    2024 arXiv

  36. [44]

    General parameter-shift rules for quantum gradients,

    D. Wierichs, J. Izaac, C. Wang, and C. Y.-Y. Lin, “General parameter-shift rules for quantum gradients,” Quantum, vol. 6, p. 677, 2022. [Online]. Available: https://doi.org/10.22331/q-2022-03-30-677

    2022 doi

  37. [45]

    S. M. Barnett and P. M. Radmore,Methods in Theoreti- cal Quantum Optics. Oxford: Oxford University Press, 2002

    2002

  38. [46]

    Ma- trix decompositions in quantum optics: Takagi/autonne, bloch–messiah/euler, iwasawa, and williamson,

    M. Houde, W. McCutcheon, and N. Quesada, “Ma- trix decompositions in quantum optics: Takagi/autonne, bloch–messiah/euler, iwasawa, and williamson,”Cana- dian Journal of Physics, vol. 102, no. 10, pp. 497–507, 2024

    2024

  39. [47]

    Further compactifying linear optical unitaries,

    B. A. Bell and I. A. Walmsley, “Further compactifying linear optical unitaries,”APL Photonics, vol. 6, no. 7, 2021

    2021

  40. [48]

    Arbi- trary optical wave evolution with fourier transforms and phase masks,

    V. L´ opez-Pastor, J. Lundeen, and F. Marquardt, “Arbi- trary optical wave evolution with fourier transforms and phase masks,”Optics Express, vol. 29, pp. 38 441–38 450, 2021

    2021

  41. [49]

    Near-optimal decomposi- tion of unitary matrices using phase masks and the dis- crete fourier transform,

    V. Girouard and N. Quesada, “Near-optimal decomposi- tion of unitary matrices using phase masks and the dis- crete fourier transform,”Journal of the Optical Society of America B, vol. 43, no. 3, pp. A66–A73, 2026

    2026

  42. [50]

    Single-photon detectors for optical quantum information applications,

    R. H. Hadfield, “Single-photon detectors for optical quantum information applications,”Nature Photonics, vol. 3, no. 12, pp. 696–705, 2009. [Online]. Available: https://doi.org/10.1038/nphoton.2009.230

    2009 doi

  43. [51]

    Invited review article: Single-photon sources and detectors,

    M. D. Eisaman, J. Fan, A. Migdall, and S. V. Polyakov, “Invited review article: Single-photon sources and detectors,”Review of Scientific Instruments, vol. 82, no. 7, p. 071101, 2011. [Online]. Available: https://doi.org/10.1063/1.3610677

    2011 doi

  44. [52]

    Accurate un- supervised photon counting from transition-edge-sensor signals,

    N. Dalbec-Constant, G. Thekkadath, D. England, B. Sussman, T. Gerrits, and N. Quesada, “Accurate un- supervised photon counting from transition-edge-sensor signals,”Phys. Rev. Appl., vol. 24, no. 3, p. 034018, 2025

    2025

  45. [53]

    Training gaussian boson sampling distributions,

    L. Banchi, N. Quesada, and J. M. Arrazola, “Training gaussian boson sampling distributions,”Phys. Rev. A, vol. 102, no. 1, p. 012417, 2020

    2020

  46. [54]

    Inverting modified matrices,

    M. A. Woodbury, “Inverting modified matrices,” Statis- tical Research Group, Princeton University, Princeton, NJ, Memorandum Report 42, 1950

    1950

  47. [55]

    Gaussian boson sampling using threshold detectors,

    N. Quesada, J. M. Arrazola, and N. Killoran, “Gaussian boson sampling using threshold detectors,”Phys. Rev. A, vol. 98, no. 6, p. 062322, 2018

    2018

  48. [56]

    Photon-number moments and cumulants of gaussian states,

    Y. Cardin and N. Quesada, “Photon-number moments and cumulants of gaussian states,”Quantum, vol. 8, p. 1521, 2024

    2024

  49. [57]

    Strong simulation of linear optical processes,

    N. Heurtel, S. Mansfield, J. Senellart, and B. Valiron, “Strong simulation of linear optical processes,”Computer Physics Communications, vol. 291, p. 108848, 2023

    2023

  50. [58]

    Asymptotically fault-tolerant programmable photon- 14 ics,

    R. Hamerly, S. Bandyopadhyay, and D. Englund, “Asymptotically fault-tolerant programmable photon- 14 ics,”Nature Communications, vol. 13, 2022. [Online]. Available: https://doi.org/10.1038/s41467-022-34308-3

    2022 doi

  51. [59]

    Fast prediction of phase shifters phase error distributions using gaussian processes,

    Z. Zhang, “Fast prediction of phase shifters phase error distributions using gaussian processes,”arXiv preprint arXiv:2504.05685, 2025. [Online]. Available: https://doi.org/10.48550/arXiv.2504.05685

    2025 doi arXiv

  52. [60]

    Perceval: A Software Platform for Discrete Variable Photonic Quantum Computing,

    N. Heurtel, A. Fyrillas, G. d. Gliniasty, R. Le Bihan, S. Malherbe, M. Pailhas, E. Bertasi, B. Bourdoncle, P.-E. Emeriau, R. Mezher, L. Music, N. Belabas, B. Valiron, P. Senellart, S. Mansfield, and J. Senellart, “Perceval: A Software Platform for Discrete Variable Photonic Qu...

    2023

  53. [61]

    Measurement of subpicosecond time intervals between two photons by interference,

    C. K. Hong, Z. Y. Ou, and L. Mandel, “Measurement of subpicosecond time intervals between two photons by interference,”Phys. Rev. Lett., vol. 59, no. 18, pp. 2044– 2046, 1987

  54. [62]

    Exponentially- improved effective descriptions of physical bosonic sys- tems,

    V. Upreti, N. Quesada, and U. Chabaud, “Exponentially- improved effective descriptions of physical bosonic sys- tems,”arXiv preprint arXiv:2604.18720, 2026

    2026 arXiv

  55. [2026]

    Available: https://doi.org/10.1038/ s41586-026-10262-8

    [Online]. Available: https://doi.org/10.1038/ s41586-026-10262-8

Pith tools

Reviewed August 2, 2026 · model on record in the stance chip above.