Pith. sign in

REVIEW 5 major objections 5 minor 40 references

The paper shows quantum ghost imaging and Grover's search algorithm are the same operation: oracle phase-marking on one photon and a diffusion measurement on its entangled partner.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

Quantum ghost imaging with Hadamard measurement patterns reproduces one iteration of Grover's search, including a condition (database size ≤ half the pixel space) for clean extraction of the marked element.

T0 review reviewed 2026-08-05 challenge →

load-bearing objection A genuinely interesting conceptual bridge between ghost imaging and Grover search, with a new separation condition, but the experimental claim rests on an idealized bucket-detector model that the setup does not quite satisfy. the 5 major comments →

arxiv 2508.11296 v1 pith:IYAKTWX3 submitted 2025-08-15 quant-ph physics.optics

Unveiling the link between quantum ghost imaging and Grover's quantum searching algorithm

Neelan Gounden , Fazilah Nothlawala , Paola C. Obando , Thomas Konrad , Andrew Forbes , Isaac Nape This is my paper
classification quant-ph physics.optics
keywords quantum ghost imagingGrover's search algorithmspatial entanglementpixel-state encodingHadamard basisphase imagingbucket detectionChoi-Jamiolkowski isomorphism
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The reading

The paper claims that quantum ghost imaging and Grover's quantum search algorithm are the same kind of operation, performed in different hardware. In ghost imaging, a phase object is encoded on one entangled photon and the partner photon is measured with patterned masks; in Grover's algorithm, an oracle marks database entries and a diffusion step amplifies them. The authors prove the mathematical equivalence by showing that the Grover diffusion operator can be absorbed into the ghost-imaging measurement basis, and they verify it experimentally with spatial light modulators and entangled photon pairs. If correct, this means a camera-like phase measurement can perform the search step of Grover's algorithm without a separate amplification circuit, and it gives a concrete condition: marked elements appear as positive-amplitude pixels distinct from an inverted unmarked background when the database occupies at most half of the full Hilbert space.

Core claim

The central claim is that bucket-detecting the signal photon in a ghost-imaging setup produces exactly the oracle-marked search state on the idler, $|\Psi_c\rangle = \sum_i c_i o_i |i\rangle$, matching the state inside Grover's algorithm after the oracle call. Consequently, measuring the idler in the basis $\hat{D}|i\rangle = \frac{2}{\sqrt{N}}|s_0\rangle - |i\rangle$ reproduces Grover's diffusion step and yields the same detection probabilities as one Grover iteration. For ghost imaging with Hadamard-adjusted masks $|b_i\rangle = (|\tilde{0}\rangle + |h_i\rangle)/\sqrt{2}$, the reconstructed image decomposes into solution and non-solution subspaces that are disjoint exactly when the databas

What carries the argument

The carrying object is the bucket-detector transfer followed by diffusion-absorbing projective masks. The bucket detector (a single-mode fiber plus avalanche photodiode) projects the signal photon and transfers the oracle coefficients $c_i$ onto the idler amplitudes, creating $|\Psi_c\rangle$. The Grover diffusion operator $\hat{D} = 2|s_0\rangle\langle s_0| - I$ is then not applied to the idler state but absorbed into the measurement basis states $\hat{D}|i\rangle$, or into Hadamard-adjusted masks $(|\tilde{0}\rangle + |h_i\rangle)/\sqrt{2}$, so that coincidence probabilities directly express the reflection of amplitudes about their mean. The formal link between sequential oracle-plus-diffu

Load-bearing premise

The bucket detector is assumed to transfer the oracle phase coefficients onto the idler amplitudes cleanly; because the single-mode fiber accepts a Gaussian mode, this requires the overlap between the fiber mode and each pixel state to be uniform or absorbed into the coefficients, otherwise the marking is distorted and the Grover equivalence breaks down.

What would settle it

Measure the idler amplitude for a single marked pixel as a function of its transverse position within the SPDC beam: if the Gaussian-mode overlap $\langle g|i\rangle$ varies from pixel to pixel, the recovered marked-element amplitudes will be suppressed away from the beam center and the contrast threshold will shift from $n = N/2$. Replacing the bucket detector with a mode-resolving measurement and showing that the predicted state $|\Psi_c\rangle$ is not produced would also settle the claim.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • A single ghost-imaging measurement in the Grover-projection basis yields the same detection probabilities as one iteration of Grover's algorithm, so the diffusion step does not need to be applied to the photon state.
  • When the database occupies at most half of the pixel Hilbert space, the reconstructed image separates cleanly: marked elements have positive amplitudes and unmarked elements are inverted, so the searched item is read out directly as bright pixels.
  • Using Hadamard-adjusted masks projects many pixels at once, avoiding single-pixel noise and scaling to larger Hilbert spaces, as demonstrated at dimensions 32, 64, and 128.
  • The equivalence holds for multiple marked elements, since the oracle encodes arbitrary sets of pixels with $\pi$ phases rather than a single item.
  • The authors claim this gives ghost imaging a computational advantage over multi-qubit search implementations for this task.

Where Pith is reading between the lines

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

  • Inference: because the protocol realizes only the first Grover iteration in the image plane, iterating the procedure or cascading masks should correspond to repeated Grover iterations and could improve contrast beyond the $n \le N/2$ threshold; this is a testable extension.
  • Inference: the Choi–Jamiolkowski-style mapping suggests that other single-iteration quantum circuits, such as amplitude-estimation variants, could be implemented as ghost-imaging measurements with suitably engineered entangled spatial modes.
  • Inference: the Gaussian-overlap caveat predicts a position-dependent distortion of marked-pixel amplitudes; shaping the pump or the fiber mode could restore uniform oracle transfer and extend the equivalence to larger databases.
  • Inference: the $n \le N/2$ separation condition has a direct experimental translation for Gaussian SPDC beams: the object or database size should stay within roughly half the beam's active pixel footprint, giving a practical alignment rule for quantum search by imaging.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 5 minor

Summary. The paper claims a conceptual and operational link between quantum ghost imaging (GI) and Grover's quantum search algorithm (GSA). The authors consider entangled photon pairs in a high-dimensional pixel basis, encode an 'oracle' via phase objects on the signal photon, and argue that a bucket-detector click transfers the oracle coefficients onto the idler, yielding the state |Ψ_c⟩ = Σ_i a_i c_i |i⟩. They show that measuring the idler in a 'Grover basis' D|i⟩ reproduces Grover's detection probabilities, and that a Hadamard-superposition basis yields a ghost image in which marked elements have positive amplitudes and unmarked elements negative amplitudes. A separation condition n ≤ d/2 is derived, below which the solution and non-solution subspaces do not overlap. Experiments with a fixed Hilbert-space dimension d = 8 and database sizes n = 2,4,6,8, plus reconstructions of objects at d = 32,64,128, are presented as supporting evidence.

Significance. If the central mapping is correct, the paper offers an interesting unification of quantum imaging and quantum search, and demonstrates experimentally that a ghost-imaging measurement with suitably chosen masks can highlight oracle-marked elements. The high-dimensional pixel encoding and the simultaneous (tensor-product) action in GI, contrasted with the sequential (consecutive) action in GSA, is a thought-provoking use of the Choi-Jamiolkowski perspective. The independent Hadamard-basis derivation is a useful contribution. However, the bridge between the actual bucket detector and Eq. (3) is idealized, and the claimed computational advantage is not substantiated. With careful revisions, the manuscript could be made sound, but at present these load-bearing points need work.

major comments (5)
  1. [Theory, Eq. (3) and Experimental setup] The clean transfer in Eq. (3) is not what a single-mode-fiber bucket detector implements. The SMF accepts a Gaussian mode |g⟩, so the heralded idler state is proportional to Σ_i a_i c_i ⟨g|i⟩ |i⟩ rather than Σ_i a_i c_i |i⟩. In the experiment the 3 mm beam is demagnified into a 4.4 μm-core fiber, and the super-pixel sizes change (8/15/30 SLM pixels), so ⟨g|i⟩ is pixel-dependent. Unless this overlap is uniform over the database or is calibrated into the amplitudes, the Grover probabilities in Eq. (8) and the reconstruction in Eq. (10) do not follow from the actual setup. This is load-bearing for the n ≤ d/2 separation condition.
  2. [Theory, Eq. (10)] Eq. (10) is stated without derivation, yet it is the basis for decomposing the reconstructed image into solution and non-solution subspaces and for the experimental reconstructions. The authors should provide a step-by-step derivation from Eqs. (8)–(9), fully define tilde{c}_i, and state where the Gaussian coefficients a_i enter and exit. This is not a cosmetic issue: the subspace-separation claim rests on this equation.
  3. [Results: computational advantage claim] The statement that 'GI offers a computational advantage compared to multi-qubit search algorithms' is unsupported. The presented protocol requires measuring all H = M^2 projective masks; for d = 128 this is 16,384 coincidence measurements at 2 s integration time, whereas Grover's algorithm uses O(√N) oracle queries. If the intended advantage is that high-dimensional pixel states require fewer physical particles than many qubits, that should be stated and formalized. As written, this claim is misleading and needs to be removed or substantially qualified.
  4. [Results, Figs. 5 and 6] The experimental evidence for the central condition n ≤ d/2 is presented without error bars, coincidence-count uncertainties, or a quantitative contrast/overlap metric. The distinction between 'no overlap' (n = 2,4) and 'overlap' (n = 6,8) is made by visual inspection. Please add quantitative measures such as marked-to-unmarked amplitude ratio, an overlap integral, or signal-to-background ratio, with uncertainties, to support the claimed transition.
  5. [Theory, Eq. (6)] The single-pixel 'Grover basis' equivalence is partly circular: the measurement states D|i⟩ are chosen precisely so that the probabilities in Eq. (5) reproduce Grover's distribution. This should be acknowledged as an equivalence by construction for that basis. The independent content is the Hadamard-basis derivation in Eqs. (7)–(10). A clarifying sentence would prevent overclaiming and help the reader understand what is genuinely derived versus defined.
minor comments (5)
  1. [Notation, Eq. (5)] The expression for P_l in Eq. (5) is hard to parse; ensure the absolute-value squared and normalization are explicit, and check consistency between symbols c_i and a_i in Eqs. (2)–(3).
  2. [Notation, d vs M] The symbol d is used both for the Hilbert-space dimension (Fig. 2) and for the side length of the mask grid (Fig. 6). Use a separate symbol, e.g. M, for the mask side length to avoid confusion.
  3. [Eq. (12)] The index range for the projective masks is unclear ('j ∈ [0,(H-M)-1]'). Define H and the total number of masks explicitly.
  4. [References] Reference [33] appears to be about a different topic than photonic integrated circuits for Grover's algorithm; please verify the intended citation.
  5. [Figure 6] The insets showing theoretical predictions in Figs. 6(e)–(g) lack a legend or a fit metric. Please state what is plotted and how 'good agreement' is quantified.

Circularity Check

1 steps flagged

Partial circularity: the 'Grover projection basis' route is defined as Grover's diffusion, so that equivalence is built in; the Hadamard-basis route is independent.

specific steps
  1. self definitional [Theory and Concept, Equations (5)-(6)]
    "instead of applying the diffusion operator to the state, |Ψ_c⟩, we can absorb it into the measurement basis states, |i⟩, resulting in the unnormalised measurement states, \hat D†|i⟩ = 2/√d |0⟩ − |i⟩ (6) ... In this perspective, Equation 5 measures the probability of detecting the marked element in the idler (search workspace) given that it has been marked in the signal (oracle workspace) photon."

    The GI projective basis in Eq. (6) is defined as \hat D†|i⟩, i.e., as Grover's diffusion operator acting on the computational basis. Under Born's rule the resulting probability is exactly P_i = |⟨i|\hat D|Ψ_c⟩|^2, which is Grover's Eq. (5). Therefore the claimed match between GI and GSA in this branch is an identity inserted by the definition of the measurement masks, not a consequence derived from the GI setup. The Hadamard-basis analysis (Eqs. 7–10) and the subspace-separation condition are independent and non-circular, so the circularity is partial.

full rationale

Most of the derivation chain is self-contained: Eqs. (1)-(2) define the entangled state and oracle; Eq. (3) is an idealization of the bucket transfer (a physical/correctness assumption, not a circular one); Eqs. (7)-(10) are a straightforward Hadamard-basis calculation leading to Eq. (10) and the n ≤ d/2 separation condition, and the experiments test these predictions. The main circular step is Eq. (6): because the 'Grover projection basis' is defined as \hat D†|i⟩, the probabilities in Eq. (5) are Grover probabilities by construction. This is an explicit reformulation rather than a hidden fit, and the paper has independent content, but it does mean part of the claimed unification is built into the choice of measurement basis. The self-citations ([30], [32], [40]) are contextual or technical and not load-bearing; no uniqueness claim is imported from the authors' prior work. Score 4 reflects one definitional reduction alongside genuinely independent Hadamard-basis results and experimental evidence.

Axiom & Free-Parameter Ledger

0 free parameters · 5 axioms · 0 invented entities

The central claim rests on standard quantum mechanics and the assumption that bucket detection faithfully transfers oracle phases to the idler. No fitted parameters and no new entities are introduced.

axioms (5)
  • domain assumption The bucket detector projects the signal photon onto a mode whose pixel-space representation is uniform (or absorbed into the coefficients c_i), so that the heralded idler state is Σ_i c_i o_i |i⟩.
    Theory section, Eq. 3: the bucket detector is a single-mode fiber that accepts a Gaussian mode. The paper states this 'transfers the coefficients c_i from the signal to the idler arm' without modeling the Gaussian overlap ⟨g|i⟩. If the overlap varies, the oracle marking is not faithfully transferred.
  • domain assumption The SPDC state has Schmidt rank at least d and the pixel states {|i⟩} form an orthonormal basis.
    Standard for high-dimensional position entanglement; the paper cites Ref [35] for the 549-dimensional pixel space. Needed for the state in Eq. (1).
  • domain assumption The object/oracle is a pure phase object with phase shifts of 0 or π, i.e., o_i ∈ {+1,-1}.
    Used throughout to define the marking operator Ô. In the experiment, the SLM encodes the object with a phase grating, but the image reconstruction assumes binary phase.
  • standard math The Walsh-Hadamard masks in Eq. (11) form an orthogonal basis over the pixel space, and the measurement basis in Eq. (12) is normalized as a superposition of the first mask and each other mask.
    Standard property of Walsh-Hadamard functions; used to derive Eq. (10).
  • standard math The Choi-Jamiolkowski isomorphism equates consecutive operations on a single system with simultaneous operations on entangled systems.
    Invoked explicitly in the text to justify the tensor product vs product of operators.

reviewed 2026-08-05 · how reviews work

0 comments
Cite this review

Pith. "Pith review of Unveiling the link between quantum ghost imaging and Grover's quantum searching algorithm." pith.science (2026). https://pith.science/paper/IYAKTWX3

@misc{pith2026250811296,
  author       = {Pith},
  title        = {Pith review of: Unveiling the link between quantum ghost imaging and Grover's quantum searching algorithm},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IYAKTWX3}},
  note         = {Machine review of arXiv:2508.11296}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Photonic quantum technologies have become pivotal in the implementation of communication, imaging and computing modalities. Among these applications, quantum ghost imaging (GI) exploits photon correlations to surpass classical limits, with recent advances in spatial-mode encoding and phase imaging. In parallel, all-optical computing offers powerful, passive-light processing capabilities. Here, we explore the intersection of these domains, revealing a conceptual and operational link between GI and Grover's quantum search algorithm (GSA) which is designed to search for elements in an unstructured database. Here, the elements are encoded as phases in the position basis states of photons. To show this, we use entangled photon pairs, with one photon encoding the oracle features while the other photon is used to find the marked element.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

40 extracted references · 40 canonical work pages

  1. [1]

    Forbes, M

    A. Forbes, M. Youssef, S. Singh, I. Nape, B. Ung, Applied Physics Letters2024, 124, 11

    work page

  2. [2]

    Yin, Y.-H

    J. Yin, Y.-H. Li, S.-K. Liao, M. Yang, Y. Cao, L. Zhang, J.-G. Ren, W.-Q. Cai, W.-Y. Liu, S.-L. Li, et al., Nature 2020, 582, 7813 501

    work page 2020

  3. [3]

    Liao, W.-Q

    S.-K. Liao, W.-Q. Cai, W.-Y. Liu, L. Zhang, Y. Li, J.-G. Ren, J. Yin, Q. Shen, Y. Cao, Z.-P. Li, et al., Nature 2017, 549, 7670 43

    work page 2017

  4. [4]

    J. L. O’brien, Science 2007, 318, 5856 1567. 11

    work page 2007

  5. [5]

    Zhong, H

    H.-S. Zhong, H. Wang, Y.-H. Deng, M.-C. Chen, L.-C. Peng, Y.-H. Luo, J. Qin, D. Wu, X. Ding, Y. Hu, et al., Science 2020, 370, 6523 1460

    work page 2020

  6. [6]

    T. D. Ladd, F. Jelezko, R. Laflamme, Y. Nakamura, C. Monroe, J. L. O’Brien, nature 2010, 464, 7285 45

    work page 2010

  7. [7]

    Nimbe, B

    P. Nimbe, B. A. Weyori, A. F. Adekoya, Quantum Information Processing2021, 20, 2 80

    work page

  8. [8]

    Rist` e, M

    D. Rist` e, M. P. Da Silva, C. A. Ryan, A. W. Cross, A. D. C´ orcoles, J. A. Smolin, J. M. Gambetta, J. M. Chow, B. R. Johnson, npj Quantum Information2017, 3, 1 16

    work page

  9. [9]

    Gilaberte Basset, F

    M. Gilaberte Basset, F. Setzpfandt, F. Steinlechner, E. Beckert, T. Pertsch, M. Gr¨ afe, Laser & Pho- tonics Reviews 2019, 13, 10 1900097

    work page 2019

  10. [10]

    Moreau, E

    P.-A. Moreau, E. Toninelli, T. Gregory, M. J. Padgett, Nature Reviews Physics2019, 1, 6 367

    work page

  11. [11]

    Moodley, A

    C. Moodley, A. Forbes, JOSA B 2023, 40, 12 3073

    work page 2023

  12. [12]

    Gregory, P.-A

    T. Gregory, P.-A. Moreau, E. Toninelli, M. J. Padgett, Science advances 2020, 6, 6 eaay2652

    work page 2020

  13. [13]

    Moreau, E

    P.-A. Moreau, E. Toninelli, T. Gregory, M. J. Padgett, Laser & Photonics Reviews2018, 12, 1 1700143

    work page

  14. [14]

    T. B. Pittman, Y. H. Shih, D. V. Strekalov, A. V. Sergienko, Phys. Rev. A1995, 52 R3429

    work page

  15. [15]

    A. F. Abouraddy, B. E. A. Saleh, A. V. Sergienko, M. C. Teich, Phys. Rev. Lett.2001, 87 123602

    work page 2001

  16. [16]

    B. I. Erkmen, J. H. Shapiro, Advances in Optics and Photonics2010, 2, 4 405

    work page

  17. [17]

    Zhang, Z

    Y. Zhang, Z. He, X. Tong, D. C. Garrett, R. Cao, L. V. Wang, Science Advances2024, 10, 10 eadk1495

    work page

  18. [18]

    Ryczkowski, M

    P. Ryczkowski, M. Barbier, A. T. Friberg, J. M. Dudley, G. Genty, Nature Photonics 2016, 10, 3 167

    work page 2016

  19. [19]

    B. Jack, J. Leach, J. Romero, S. Franke-Arnold, M. Ritsch-Marte, S. Barnett, M. Padgett, Physical review letters 2009, 103, 8 083602

    work page 2009

  20. [20]

    R. S. Aspden, D. S. Tasca, R. W. Boyd, M. J. Padgett, New Journal of Physics2013, 15, 7 073032

    work page

  21. [21]

    Shih, IEEE Journal of Selected Topics in Quantum Electronics2007, 13, 4 1016

    Y. Shih, IEEE Journal of Selected Topics in Quantum Electronics2007, 13, 4 1016

    work page

  22. [22]

    Defienne, W

    H. Defienne, W. P. Bowen, M. Chekhova, G. B. Lemos, D. Oron, S. Ramelow, N. Treps, D. Faccio, Nature Photonics 2024, 18, 10 1024

    work page 2024

  23. [23]

    M. J. Padgett, R. W. Boyd, Philosophical Transactions of the Royal Society A: Mathematical, Physi- cal and Engineering Sciences2017, 375, 2099 20160233

    work page 2099

  24. [24]

    Moodley, A

    C. Moodley, A. Forbes, Laser & Photonics Reviews2024, 18, 8 2300939

    work page

  25. [25]

    Ferri, D

    F. Ferri, D. Magatti, A. Gatti, M. Bache, E. Brambilla, L. A. Lugiato, Physical review letters2005, 94, 18 183602

    work page

  26. [26]

    R. S. Bennink, S. J. Bentley, R. W. Boyd, Physical review letters2002, 89, 11 113601

    work page

  27. [27]

    Bornman, S

    N. Bornman, S. Prabhakar, A. Vall´ es, J. Leach, A. Forbes, New Journal of Physics2019, 21, 7 073044

    work page

  28. [28]

    L. Wang, S. Zhao, Photonics Research2016, 4, 6 240

    work page

  29. [29]

    Zhang, X

    Z. Zhang, X. Ma, J. Zhong, Nature communications2015, 6, 1 6225

    work page

  30. [30]

    Sephton, I

    B. Sephton, I. Nape, C. Moodley, J. Francis, A. Forbes, Optica 2023, 10, 2 286. 12

    work page 2023

  31. [31]

    Kwiat, J

    P. Kwiat, J. Mitchell, P. Schwindt, A. White, Journal of Modern Optics2000, 47, 2-3 257

    work page

  32. [32]

    Perez-Garcia, R

    B. Perez-Garcia, R. I. Hernandez-Aranda, A. Forbes, T. Konrad, Journal of Modern Optics2018, 65, 16 1942

    work page 1942

  33. [33]

    G. N. M. Tabia, Physical Review A2016, 93, 1 012323

    work page

  34. [34]

    L. K. Grover, In Proceedings of the twenty-eighth annual ACM symposium on Theory of computing. 1996 212–219

    work page 1996

  35. [35]

    N. H. Valencia, V. Srivastav, M. Pivoluska, M. Huber, N. Friis, W. McCutcheon, M. Malik, Quantum 2020, 4 376

    work page 2020

  36. [36]

    Moreau, E

    P.-A. Moreau, E. Toninelli, P. A. Morris, R. S. Aspden, T. Gregory, G. Spalding, R. W. Boyd, M. J. Padgett, Optics express 2018, 26, 6 7528

    work page 2018

  37. [37]

    Tasca, R

    D. Tasca, R. Aspden, P. Morris, G. Anderson, R. Boyd, M. Padgett, Optics express 2013, 21, 25 30460

    work page 2013

  38. [38]

    M. A. Nielsen, I. L. Chuang, Quantum Computation and Quantum Information: 10th Anniversary Edition, Cambridge University Press, 2010

    work page 2010

  39. [39]

    L. K. Grover, Phys. Rev. Lett.1997, 79 325

    work page 1997

  40. [40]

    Rodr ´ ıguez-Fajardo, J

    V. Rodr ´ ıguez-Fajardo, J. Pinnell, A. Forbes,Journal of Modern Optics2020, 67, 13 1176. 13

    work page

This paper was first reviewed by deepseek-v4-flash on August 5, 2026.