Pith. sign in

REVIEW 3 major objections 4 minor 2 cited by

Experimental neuromorphic computing based on quantum memristor

T0 review · 3 major / 4 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read A single-photon interferometer with measurement feedback acts as a nonlinear, memory-equipped reservoir, lowering prediction error on four tasks.

desk verdict A solid experimental proof-of-principle showing a photonic quantum memristor improves reservoir-computing performance, with a real but bridgeable gap between the analytic nonlinearity proof and the implemented feedback rules. read the letter →

arxiv 2504.18694 v2 pith:MFDF3UFG submitted 2025-04-25 quant-ph

Mirela Selimović , Iris Agresti , Micha{l} Siemaszko , Joshua Morris , Borivoje Dakić , Riccardo Albiero , Andrea Crespi , Francesco Ceccarelli
show 3 more authors
Roberto Osellame Magdalena Stobińska Philip Walther
This is my paper · ORCID
classification quant-ph
keywords quantummemristorneuromorphiccomputingreservoirsingle-photoninterferometrynonlinearitytimeseriespredictionintegratedphotonicsfeedbackloop
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

The paper aims to show that a photonic quantum memristor—a single photon passing through a tunable interferometer whose reflectivity is adjusted by previous measurement outcomes—can serve as the nonlinear, memory-bearing core of a reservoir computer, without any entangling gates. The authors report the first neuromorphic architecture built on this device and benchmark it on four tasks: reproducing monomials $x^n$ and predicting NARMA, Mackey-Glass, and Santa Fe time series. Across all tasks, enabling the memristive feedback loop lowers the prediction error compared with the same circuit with the feedback turned off, while a classical linear readout is the only trained part. The relevance is that quantum machine learning hardware can get its nonlinearity from measurement feedback rather than from hard-to-scale nonlinear optical interactions.

What carries the argument

The central object is the photonic quantum memristor realised as a tunable Mach-Zehnder interferometer with a feedback loop: the reflectivity $R_t$ of the effective beam splitter is updated from the measured output probability $p_{t,2}$ at one output mode, through a moving average or an exponential moving average. The device acts on one qubit encoded in the path degree of freedom of a single photon, with the input encoded nonlinearly into the amplitudes, for example $x \to \sqrt{x}|0\rangle + \sqrt{1-x}|1\rangle$. The feedback creates an output probability that depends on products of earlier inputs, which supplies the reservoir's nonlinearity and memory. Tracing out the feedback mode still leaves a partially coherent state, so the output can feed further quantum processing.

What would settle it

Run a noiseless numerical simulation of the exact experimental circuit with the exact feedback rules in Eqs. (2) and (4), and check whether the output probability at time $t$ contains products of earlier inputs, for example a term depending on $x_{t-1} x_t$. If the real feedback rules produce a map that is linear in past inputs once the encoding nonlinearity is factored out, then the memristor's claimed nonlinearity enhancement is absent and the performance gain should instead be attributed to the separately optimized hyperparameters.

Watch

Extended reading notes

Core claim

The central discovery is a working hybrid quantum-classical reservoir computer in which nonlinearity comes from an adaptive measurement loop rather than from entangling gates. A single photon is encoded with classical data, passed through a Mach-Zehnder interferometer whose internal phase is updated by a linear rule based on earlier detector counts, and the resulting output probabilities feed a classical linear regression. The feedback makes the transformation nonlinear in the input—already after one step the simplified model produces products like $(1-x_0)(1-x_1)$—and gives the reservoir short-term memory. The authors demonstrate that for all four tasks the memristive feedback loop improves accuracy relative to the same device without feedback, and they compare against small classical polynomial models, finding their device competitive or better on the less smooth tasks while using a single trained parameter in the time-series cases.

Load-bearing premise

The load-bearing premise is that the product-type nonlinearity proven analytically for a simplified memristor configuration—identity unitaries and the feedback rule $R_{t+1}=p_{2t}$—still holds for the more complex averaged feedback rules and optimized rotations used in the experiments; if it does not, the improvement attributed to the memristor could instead come from the separately optimised hyperparameters.

Editorial extensions

If this is right

  • The same setup, with a single tunable interferometer and feedback loop, can serve as a nonlinear activation layer for larger optical neural networks.
  • Because the feedback mode can be traced out while preserving partial coherence, multiple quantum memristors can be cascaded or networked to build more expressive quantum reservoirs.
  • The scheme requires no entangling gates, so it can run on current integrated photonic chips with classical electro-optic feedback, easing near-term scalability.
  • The comparison against small classical polynomial models suggests the quantum memristor is most useful for non-smooth time series, where local polynomial approximations fail.
  • Only a linear regression is trained, so the approach inherits the low training cost of reservoir computing while adding a quantum-generated nonlinearity.

Reading between the lines

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

  • If the feedback-generated product nonlinearity scales as the simplified derivation indicates, similar measurement-feedback devices in other platforms, such as circuit-QED or cavity-QED systems, could generate comparable nonlinearities without entangling gates.
  • A testable extension would be to run the same memristor reservoir on a task whose target depends on higher-order products of past inputs, such as predicting $x_t x_{t-1}$, where the claimed product nonlinearity should yield a clear advantage over a no-feedback baseline.
  • The nonlinearity argument is worked out for one simplified configuration; a full derivation for the experimental averaged feedback rules would make the attribution of the performance gain to the memristor mechanism more robust, since the hyperparameters are optimised separately in the with- and without-feedback cases.
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 / 4 minor

Summary. The paper reports an experimental implementation of reservoir computing based on a photonic quantum memristor: single photons are encoded into path-encoded qubit states, processed by an integrated three-mode interferometer whose central beam-splitter reflectivity is updated by a feedback rule depending on previous measurement outcomes, and then read out by a classical linear regression. The scheme is benchmarked on four tasks: reproduction/extrapolation of monomial functions x^n, and one-step-ahead prediction of the NARMA, Mackey-Glass, and Santa Fe time series. In each task the authors compare the memristive-feedback circuit against the same circuit with the feedback loop disabled, reporting consistently lower mean squared error with feedback, and they also compare against simple classical polynomial models. The central claim is that the memristive feedback loop enhances the achievable nonlinearity and thereby improves the performance of the tested algorithms, constituting a first neuromorphic architecture based on a photonic quantum memristor.

Significance. If the central claim is established, the work is significant: it demonstrates that a single-photon Mach-Zehnder interferometer with a measurement-dependent reflectivity can serve as a nonlinear, memory-equipped reservoir for quantum reservoir computing, without entangling gates and on an integrated photonic platform. The direct experimental with-versus-without-feedback control is a genuine and valuable design, and the supplementary material provides useful supporting material, including on-chip tomography showing partial coherence of the output state, a comparison with numerical simulations, and a comparison with classical benchmark models. The paper is also commendably explicit about which parameters are optimized. However, the analytical bridge between the simplified model in which the feedback nonlinearity is proven and the actual implemented feedback rules is incomplete, and some of the headline comparisons mix experimental results with simulated baselines; these issues need to be addressed before the performance claims can be accepted as stated.

major comments (3)
  1. [Supplementary Note II (Eqs. S1-S8 and Eq. S14)] The analytical argument that the feedback loop produces product-type nonlinearity is derived only for U1 = U3 = I and for the simplified rule R_{t+1} = p_{2t}, yielding outputs such as (1-x0)(1-x1). The experiments, however, use the moving-average rule in Eq. (2) of the main text or the recursive exponential-moving-average rule in Eq. (4), and the recursive solution in Eq. (S14) of the Supplement makes R_t depend on all past probabilities with exponentially decaying weights. The supplement asserts that the general setting still contains product-type nonlinear terms, but no derivation is given for the implemented feedback rules. Since the paper's central causal claim is that the memristive feedback specifically enhances nonlinearity, this missing bridge is load-bearing. Please provide an explicit derivation, or a numerical demonstration, that the output nonlinearity for the actual feedback rules and unitaries has the same functional structure as the simplified product form, rather than merely being a different nonlinearity that happens to help on the tested tasks.
  2. [Fig. 3 caption and Supplementary Note IIA] For the monomial task, the rotations U1 and U2 and the feedback coefficients a and b are optimized separately for the with-feedback and without-feedback arms. This means the performance gain attributed to the memristor is not isolated from the gain due to the additional tunable parameters available in the feedback arm. The no-feedback arm does not have the same number of free parameters, so the observed improvement is an upper bound on the effect of the feedback nonlinearity itself. Please report a matched-budget comparison, for example by fixing U1 and U2 to the values optimized for the feedback case and only removing the feedback loop, or by giving the no-feedback arm the same number of optimized parameters in a controlled way.
  3. [Table I and Supplementary Note IIC] The comparison between QMEM and the classical polynomial models in Table I is not apples-to-apples: the QMEM row is an experimental result, while the classical rows are numerical simulations. Supplementary Table S2 shows that experimental noise is task-dependent, with the Santa Fe experimental MSE (about 2.3e-2) being substantially worse than the simulated value (about 9.2e-3), so a single experimental point cannot be directly compared with simulated classical baselines without conflating device noise with model capability. Please provide either classical baselines run on the same experimental hardware, or a simulation-level comparison for both QMEM and the classical models, with matching uncertainty estimates.
minor comments (4)
  1. [Supplementary Fig. S8 and main text, Section II] The main text states that for the time-series tasks the only hyperparameter is the memory decay md, but for the monomial task U1, U2, a, and b are also optimized; Table I's claim that the model 'features only 1 free variable' is therefore true only for the time-series tasks and should be stated as such.
  2. [Notation, Section II and Supplementary Note II] The notation for the unitaries is inconsistent: the main text uses U1, Umem, and U2 for the three interferometer stages, while the supplement uses U1, U2, and U3, with U2 denoting the memristor unitary. Please harmonize the notation to avoid confusion.
  3. [Fig. 4 and Supplementary Table S2] The main-text figures show a single representative run for each task, while the supplement reports averages over only three runs; adding error bars or confidence intervals to the main figures would make the reported differences between the with- and without-feedback cases more convincing, especially where the margins are small.
  4. [Supplementary Note IIC, Fig. S9] The statement that 'the quantum model is superior for time series which are not smooth' is a post-hoc interpretation based on three datasets; the paper should either hedge this claim or support it with a broader family of tasks.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the feedback nonlinearity is derived from the circuit model and the with/without-feedback comparison is a genuine control; the simplified derivation for Eq. (4) is a proof gap, not circularity.

full rationale

The paper's central claim is that the memristive feedback loop adds nonlinearity and thereby improves performance. The analytical support in Supplementary Note II computes output probabilities for a simplified feedback rule directly from the Mach-Zehnder unitary, yielding product-type nonlinear terms such as (1-x0)(1-x1). This is a concrete circuit-level calculation, not an input that is renamed as the conclusion. The actual experimental feedback rule in Eq. (4) is not fully covered by that simplified derivation, but that is an unproven generalization and a rigor concern, not a circular reduction: the paper does not define the feedback rule in terms of the target MSE, nor does it fit a parameter and then relabel it as a prediction. Hyperparameters are optimized separately for the with- and without-feedback arms and for each task, which is standard practice and makes the comparison fairer rather than forcing the reported improvement. The comparison with the no-feedback case is a genuine control, and the comparison with classical polynomial models provides external benchmarking. Citations to prior work on the photonic quantum memristor contain overlapping authors, but that prior work is an independent experimental demonstration and the present supplementary note re-derives the output-state dynamics and purity from the circuit model; the self-citation is not load-bearing. No step in the paper's derivation chain reduces, by definition or by fitted-parameter renaming, to its own inputs.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central experimental claim is not circular: the feedback nonlinearity is derived analytically and the with/without feedback comparison is a direct control. However, the claim that the memristor is essential rests on fitted hyperparameters and a simplified analytic model that may not capture the full dynamics. The free parameters are honest but undercounted in the resource comparison.

free parameters (4)
  • U1, U2 rotations (monomial task) = optimized numerically, values not reported
    Chosen by gradient descent to minimize p_D - x^n for each n (Supplementary Note IIA). These unitaries set the reservoir features and are re-optimized separately for the no-feedback baseline.
  • feedback coefficients a, b = not reported
    In Eq. (2), R_t = (1/m) sum_{T=t-m}^{t} (a p_{T,2} + b). These coefficients control how strongly the measurement outcome changes the interferometer reflectivity and are optimized per monomial degree.
  • memory extent m / memory decay md = md = 2 (Mackey-Glass), md = 6 (Santa Fe), m = 4 or 6 (NARMA)
    Hyperparameters setting the timescale of the moving or exponential average in the feedback rule. md is selected from the MSE versus memory-extent scan in Fig. S8, which uses the data to pick the best value.
  • linear readout weights = trained via linear regression
    The readout yout = sum_i y_i w_i is the only trained part. The paper's Table I claims 'only 1 free variable', but the readout has multiple weights (one per output mode plus bias), undercounting the trained parameters.
assumptions (4)
  • domain assumption The integrated photonic circuit implements the stated unitaries U_enc, U1, Umem, U2 after calibration.
    All results depend on the chip's calibrated phase shifters and the Reck decomposition. Mis-calibration would change the reservoir features. See Section III and Supplementary Note III.
  • domain assumption The simplified analytical nonlinearity derivation extends to the actual feedback rules in Eqs. (2) and (4).
    The explicit product form x_t x_{t-1} is shown only for U1=U3=I and R_{t+1}=p_{2t} in Supplementary Note II. The experimental rules are more complex, so the stated mechanism is an extrapolation.
  • domain assumption Post-selecting on heralded single photons yields unbiased estimates of the ideal output probabilities.
    Loss and dark counts are not modeled in the comparisons. The experimental probabilities are computed as N_i,a / sum_j N_j,a from coincidence counts, which assumes the detection statistics faithfully represent the ideal single-photon probabilities.
  • domain assumption The four chosen tasks are representative of problems requiring nonlinearity and memory.
    The paper infers from these benchmarks that the memristor is a useful building block for neuromorphic computing; this is an inductive claim about the device's generality.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Experimental neuromorphic computing based on quantum memristor." pith.science (2026). https://pith.science/paper/MFDF3UFG

@misc{pith2026250418694,
  author       = {Pith},
  title        = {Pith review of: Experimental neuromorphic computing based on quantum memristor},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MFDF3UFG}},
  note         = {Machine review of arXiv:2504.18694}
}
read the original abstract

Machine learning has recently developed novel approaches, mimicking the synapses of the human brain to achieve similarly efficient learning strategies. Such an approach retains the universality of standard methods, while attempting to circumvent their excessive requirements, which hinder their scalability. In this landscape, quantum (or quantum inspired) algorithms may bring enhancement. However, high-performing neural networks invariably display nonlinear behaviours, which poses a challenge to quantum platforms, given the intrinsically linear evolution of closed systems. We propose a strategy to enhance the nonlinearity achievable in this context, without resorting to entangling gates and report the first neuromorphic architecture based on a photonic quantum memristor. In detail, we show how the memristive feedback loop enhances the nonlinearity and hence the performance of the tested algorithms. We benchmark our model on four tasks, a nonlinear function and three time series prediction. In these cases, we highlight the essential role of the quantum memristive element and demonstrate the possibility of using it as a building block in more sophisticated networks.

Figures

Figures reproduced from arXiv: 2504.18694 by the authors.

Figure 1
Figure 1. FIG. 1 [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Then, its internal phase is updated according to a feedback rule that depends on the measurement out￾come at one output mode, affecting the temporal dy￾namics of the device. One of its main features consists in the fact that, when tracing out the mode used for the feedback loop, the remaining reduced state may still be partially coherent relative to the joint state, as detailed in the Supplementary Material note I. … view at source ↗
Figure 3
Figure 3. FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: FIG. 4 [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

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

  1. Large-scale quantum reservoir computing using a Gaussian Boson Sampler

    quant-ph 2025-05 conditional novelty 7.0 of 10

    A frequency-multiplexed Gaussian boson sampler with over 400 modes can serve as a quantum reservoir computer, with correlation-based features and squeezed light outperforming mean-field and classical-light baselines.

  2. Coupled integrated photonic quantum memristors using a single photon source made of a colour center

    quant-ph 2026-02 conditional novelty 6.0 of 10

    Coupled photonic quantum memristors with crossed feedback, driven by a room-temperature SiV− single-photon source, show enhanced, self-intersecting hysteresis in inter-memristor input-output curves.

Reference graph

Works this paper leans on

70 extracted references · 58 canonical work pages · cited by 2 Pith papers

  1. [1]

    Abdel-Hamid, O. et al. Convolutional neural networks for speech recognition. IEEE Transactions on Audio, Speech and Language Processing 22, 1533–1545 (2014). Pub- lisher: Institute of Electrical and Electronics Engineers Inc

    work page 2014

  2. [2]

    & Radosav, D

    Bakator, M. & Radosav, D. Deep learning and medical diagnosis: A review of literature. Multimodal Technolo- gies and Interaction 2 (2018)

    work page 2018

  3. [3]

    Silver, D. et al. Mastering the game of Go without human knowledge. Nature 550, 354–359 (2017)

    work page 2017

  4. [4]

    Pierson, H. A. & Gashler, M. S. Deep learning in robotics: a review of recent research. Advanced Robotics 31, 821–835 (2017)

    work page 2017

  5. [5]

    Ibm’s deep blue chess grandmaster chips

    Hsu, F.-H. Ibm’s deep blue chess grandmaster chips. IEEE Micro 19, 70–81 (1999)

    work page 1999

  6. [6]

    Jumper, J. et al. Highly accurate protein structure pre- diction with AlphaFold. Nature 596, 583–589 (2021)

    work page 2021

  7. [7]

    Brown, T. B. et al. Language Models are Few-Shot Learners. arXiv preprint arXiv:2005.14165v4 (2020)

    arXiv 2020

  8. [8]

    & Hinton, G

    Krizhevsky, A., Sutskever, I. & Hinton, G. E. ImageNet classification with deep convolutional neural networks. Communications of the ACM 60, 84–90 (2017)

    work page 2017

Show all 70 references
  1. [9]

    & Ausavarung- nirun, R

    Mutlu, O., Ghose, S., G´ omez-Luna, J. & Ausavarung- nirun, R. Processing data where it makes sense: Enabling in-memory computation. Microprocessors and Microsys- tems 67, 28–41 (2019). 1903.03988

    2019 arXiv

  2. [10]

    & Kim, K

    Ham, D., Park, H., Hwang, S. & Kim, K. Neuromorphic electronics based on copying and pasting the brain. Nat Electron 4, 635–644 (2021)

    2021

  3. [11]

    Memristor-the missing circuit element

    Chua, L. Memristor-the missing circuit element. IEEE Transactions on Circuit Theory 18, 507–519 (1971)

    1971

  4. [12]

    B., Snider, G

    Strukov, D. B., Snider, G. S., Stewart, D. R. & Williams, R. S. The missing memristor found. Nature 453, 80–83

  5. [13]

    Jo, S. H. et al. Nanoscale memristor device as synapse in neuromorphic systems. Nano Lett. 10, 1297–1301 (2010)

    2010

  6. [14]

    Du, C. et al. Reservoir computing using dynamic mem- ristors for temporal information processing. Nature Com- munications 8, 2204 (2017). 8

    2017

  7. [15]

    Xiao, Y. et al. A review of memristor: material and structure design, device performance, applications and prospects. Science and Technology of Advanced Materials 24, 2162323 (2023)

    2023

  8. [16]

    Zhong, Y. et al. Dynamic memristor-based reservoir com- puting for high-efficiency temporal signal processing. Na- ture communications 12, 408 (2021)

    2021

  9. [17]

    Reck, M., Zeilinger, A., Bernstein, H. J. & Bertani, P. Experimental realization of any discrete unitary opera- tor. Physical Review Letters 73, 58–61 (1994)

    1994

  10. [18]

    Grover, L. K. A fast quantum mechanical algorithm for database search. In Proceedings of the Twenty- Eighth Annual ACM Symposium on Theory of Comput- ing, STOC ’96, 212–219 (Association for Computing Ma- chinery, New York, NY, USA, 1996)

    1996

  11. [19]

    Shor, P. W. Polynomial-time algorithms for prime factor- ization and discrete logarithms on a quantum computer. SIAM Review 41 (1999)

    1999

  12. [20]

    Quantum algorithms: An overview

    Montanaro, A. Quantum algorithms: An overview. npj Quantum Information 2 (2016). 1511.04206

    2016 arXiv

  13. [21]

    W., Hassidim, A

    Harrow, A. W., Hassidim, A. & Lloyd, S. Quantum Algo- rithm for Linear Systems of Equations. Physical Review Letters 103, 150502 (2009)

    2009

  14. [22]

    Arute, F. et al. Quantum supremacy using a pro- grammable superconducting processor. Nature 574, 505– 510

  15. [23]

    & Briegel, H

    Dunjko, V. & Briegel, H. J. Machine learning & artifi- cial intelligence in the quantum domain: a review of re- cent progress. Reports on Progress in Physics 81, 074001 (2018)

    2018

  16. [24]

    Quantum machine learning: what quantum computing means to data mining (Academic Press, 2014)

    Wittek, P. Quantum machine learning: what quantum computing means to data mining (Academic Press, 2014)

    2014

  17. [25]

    McMahon, P. L. The physics of optical computing. Na- ture Reviews Physics 5, 717–734 (2023). 2308.00088

    2023 arXiv

  18. [26]

    & Englund, D

    Hamerly, R., Bernstein, L., Sludds, A., Soljaˇ ci´ c, M. & Englund, D. Large-scale optical neural networks based on photoelectric multiplication. Physical Review X 9, 021032 (2019)

    2019

  19. [27]

    Yin, Z. et al. Experimental quantum-enhanced kernels on a photonic processor. arXiv preprint arXiv.2407.2036 (2024)

    2024

  20. [28]

    Wanjura, C. C. & Marquardt, F. Fully nonlinear neuro- morphic computing with linear wave scattering. Nature Physics 20, 1434–1440 (2024)

    2024

  21. [29]

    Suprano, A. et al. Experimental property reconstruction in a photonic quantum extreme learning machine. Phys- ical Review Letters 132, 160802 (2024)

    2024

  22. [30]

    R., Hurtado, A., P´ erez-L´ opez, D

    Rausell-Campo, J. R., Hurtado, A., P´ erez-L´ opez, D. & Capmany Francoy, J. Programmable photonic extreme learning machines. Laser & Photonics Reviews 2400870 (2024)

    2024

  23. [31]

    Mujal, P. et al. Opportunities in quantum reservoir com- puting and extreme learning machines. Advanced Quan- tum Technologies 4. 2102.11831

    arXiv

  24. [32]

    Spagnolo, M. et al. Experimental photonic quantum memristor. Nature Photonics 16, 318–323 (2022)

    2022

  25. [33]

    & Lo Franco, R

    Ferrara, A. & Lo Franco, R. Entanglement and coher- ence dynamics in photonic quantum memristors.Physical Review A 111, 012421 (2025)

    2025

  26. [34]

    The MNIST database of handwritten digit images for machine learning research [best of the web]

    Li Deng. The MNIST database of handwritten digit images for machine learning research [best of the web]. IEEE Signal Process. Mag. 29, 141–142 (2012)

    2012

  27. [35]

    & Yamamoto, N

    Kobayashi, K., Fujii, K. & Yamamoto, N. Feedback- Driven Quantum Reservoir Computing for Time-Series Analysis. PRX Quantum 5, 040325 (2024)

    2024

  28. [36]

    L., Soriano, M

    Mujal, P., Mart´ ınez-Pe˜ na, R., Giorgi, G. L., Soriano, M. C. & Zambrini, R. Time-series quantum reservoir computing with weak and projective measurements. npj Quantum Information 9, 1–10 (2023). Publisher: Nature Publishing Group

    2023

  29. [37]

    & Mujal, P

    Franceschetto, G., P/suppress lodzie´ n, M., Lewenstein, M., Ac´ın, A. & Mujal, P. Harnessing quantum back-action for time-series processing. arXiv (2024). ArXiv:2411.03979 [quant-ph]

    2024 arXiv

  30. [38]

    & Schmidhuber, J

    Hochreiter, S. & Schmidhuber, J. Long short-term mem- ory. Neural Computation 9, 1735–1780 (1997)

    1997

  31. [39]

    & Parlos, A

    Atiya, A. & Parlos, A. New results on recurrent network training: unifying the algorithms and accelerating con- vergence. IEEE Trans. Neural Netw. 11, 697–709 (2000)

    2000

  32. [41]

    & Nakajima, K

    Fujii, K. & Nakajima, K. Harnessing disordered-ensemble quantum dynamics for machine learning. Phys. Rev. Ap- plied 8, 024030 (2017)

    2017

  33. [42]

    H¨ ubner, U., Abraham, N. B. & Weiss, C. O. Dimensions and entropies of chaotic intensity pulsations in a single- mode far-infrared nh 3 laser. Physical Review A 40, 6354 (1989)

    1989

  34. [43]

    Weigend, A. S. & Gershenfeld, N. A. Results of the time series prediction competition at the santa fe institute. In IEEE international conference on neural networks , 1786– 1793 (IEEE, 1993)

    1993

  35. [45]

    & Yoshimura, K

    Inubushi, M. & Yoshimura, K. Reservoir computing be- yond memory-nonlinearity trade-off. Sci Rep 7, 10199 (2017)

    2017

  36. [46]

    Akai-Kasaya, M. et al. Performance of reservoir com- puting in a random network of single-walled carbon nan- otubes complexed with polyoxometalate. Neuromorph. Comput. Eng. 2, 014003 (2022)

    2022

  37. [47]

    Y.-C., Rastunkov, V

    Fry, D., Deshmukh, A., Chen, S. Y.-C., Rastunkov, V. & Markov, V. Optimizing quantum noise-induced reservoir computing for nonlinear and chaotic time series predic- tion. Sci Rep 13, 19326 (2023)

    2023

  38. [48]

    R., Humphreys, P

    Clements, W. R., Humphreys, P. C., Metcalf, B. J., Kolthammer, W. S. & Walmsley, I. A. Optimal design for universal multiport interferometers. Optica 3, 1460–1465 (2016)

    2016

  39. [49]

    Gattass, R. R. & Mazur, E. Femtosecond laser micro- machining in transparent materials. Nature photonics 2, 219–225 (2008)

    2008

  40. [50]

    & Osellame, R

    Corrielli, G., Crespi, A. & Osellame, R. Femtosecond laser micromachining for integrated quantum photonics. Nanophotonics 10, 3789–3812 (2021)

    2021

  41. [51]

    Ceccarelli, F. et al. Low Power Reconfigurability and Reduced Crosstalk in Integrated Photonic Circuits Fab- ricated by Femtosecond Laser Micromachining. Laser & Photonics Reviews 14, 2000024 (2020)

    2020

  42. [52]

    Mood, A. M. Introduction to the theory of statistics. (1950)

    1950

  43. [53]

    & Sciarrino, F

    Di Micco, S., Polacchi, B., Giordani, T. & Sciarrino, F. Quantum memristor with vacuum–one-photon qubits. arXiv preprint arXiv:2503.02466 (2025). 9

    2025

  44. [54]

    Carosini, L. et al. Programmable multiphoton quantum interference in a single spatial mode. Science Advances 10, eadj0993 (2024)

    2024

  45. [55]

    & Assimakopoulos, V

    Makridakis, S., Spiliotis, E. & Assimakopoulos, V. Sta- tistical and machine learning forecasting methods: Con- cerns and ways forward. PloS one 13, e0194889 (2018)

    2018

  46. [56]

    & Lukin, M

    Cong, I., Choi, S. & Lukin, M. D. Quantum convolutional neural networks. Nature Physics 15, 1273–1278 (2019)

    2019

  47. [57]

    Cimini, V. et al. Calibration of multiparameter sensors via machine learning at the single-photon level. Physical Review Applied 15, 044003 (2021)

    2021

  48. [58]

    & Petruccione, F

    Brand, D. & Petruccione, F. A quantum leaky integrate- and-fire spiking neuron and network. npj Quantum In- formation 10, 1–8 (2024)

    2024

  49. [59]

    & Sejnowski, T

    Rungratsameetaweemana, N., Kim, R., Chotibut, T. & Sejnowski, T. J. Random noise promotes slow hetero- geneous synaptic dynamics important for robust work- ing memory computation. Proceedings of the National Academy of Sciences 122, e2316745122 (2025)

    2025

  50. [60]

    Yamazaki, K., Vo-Ho, V.-K., Bulsara, D. & Le, N. Spik- ing neural networks and their applications: A review. Brain sciences 12, 863 (2022). Supplementary Material I. ON-CHIP QUANTUM TOMOGRAPHY OF QUANTUM MEMRISTOR OUTPUT ST A TE As detailed in [1] and in the main text, the pho...

    2022

  51. [61]

    Spagnolo, M. et al. Experimental photonic quantum memristor. Nature Photonics 16, 318–323 (2022). 15

    2022

  52. [62]

    Innocenti, L. et al. Potential and limitations of quantum extreme learning machin es. Communications Physics 2023 6:1 6, 1–9 (2023)

    2023

  53. [63]

    Govia, L. C. G., Ribeill, G. J., Rowlands, G. E. & Ohki, T. A . Nonlinear input transformations are ubiquitous in quantum reservoir computing. Neuromorphic Computing and Engineering 2, 014008. 2107.00147

    arXiv

  54. [64]

    An overview of gradient descent optimization algo rithms

    Ruder, S. An overview of gradient descent optimization algo rithms. arXiv preprint arXiv:1609.04747 (2016)

    2016 arXiv

  55. [65]

    Diederik, P. K. Adam: A method for stochastic optimization. Proceedings of the 3rd International Conference for Learni ng Representations, San Diego, 2015 (2015)

    2015

  56. [66]

    & Schmidhuber, J

    Hochreiter, S. & Schmidhuber, J. Long short-term memory. Neural Computation 9, 1735–1780 (1997)

    1997

  57. [67]

    & Parlos, A

    Atiya, A. & Parlos, A. New results on recurrent network training: unifying the algorithms and accelerating convergence. IEEE Trans. Neural Netw. 11, 697–709 (2000)

    2000

  58. [68]

    Shahi, S., Fenton, F. H. & Cherry, E. M. Prediction of chaotic time series using recurrent neural networks and reservoir computing techniques: A comparative study. Machine learning with applications 8, 100300 (2022)

    2022

  59. [69]

    & Nakajima, K

    Fujii, K. & Nakajima, K. Harnessing disordered-ensemble quan tum dynamics for machine learning. Phys. Rev. Applied 8, 024030 (2017)

    2017

  60. [70]

    H¨ ubner, U., Abraham, N. B. & Weiss, C. O. Dimensions and ent ropies of chaotic intensity pulsations in a single-mode far-infrared nh 3 laser. Physical Review A 40, 6354 (1989)

    1989

  61. [71]

    Weigend, A. S. & Gershenfeld, N. A. Results of the time serie s prediction competition at the santa fe institute. In IEEE international conference on neural networks , 1786–1793 (IEEE, 1993)

    1993

  62. [72]

    Mackey, M. C. & Glass, L. Oscillation and chaos in physio logical control systems. Science 197, 287–289 (1977)

    1977

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.