pith. sign in

arxiv: 2511.13774 · v3 · submitted 2025-11-15 · 🪐 quant-ph

Hybrid Predictive Quantum Feedback: Extending Qubit Lifetimes Beyond the Wiseman-Milburn Limit

Ali Abu-Nada , Aryan Iliat , Russell Ceballos This is my paper

Pith reviewed 2026-05-17 21:54 UTC · model grok-4.3

classification 🪐 quant-ph
keywords quantum feedbackqubit lifetimeWiseman-Milburn limitancilla qubithomodyne measurementpredictive controlamplitude dampingLindblad equation
0
0 comments X

The pith

A hybrid ancilla and predictor feedback scheme extends qubit lifetimes beyond the Wiseman-Milburn limit by a factor of three to four.

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

The paper introduces a hybrid quantum feedback method to reduce amplitude damping in qubits beyond what standard feedback achieves. It pairs a coherently coupled ancilla qubit that decays rapidly with a supervised predictor that forecasts the homodyne current to offset hardware delays. A Lindblad analysis supplies closed-form effective decay rates showing suppression from both the ancilla cooperativity and the prediction accuracy. Simulations with realistic parameters demonstrate roughly three to four times longer effective T1 along with better population retention. Readers care because extended qubit lifetimes directly enable more reliable quantum operations on near-term hardware.

Core claim

The central claim is that a coherently coupled fast-decaying ancilla recovers information from both field quadratures through quantum-coherent feedback while a lightweight supervised predictor phase-aligns the correction to overcome loop latency, yielding effective decay rates that are suppressed first by an ancilla cooperativity factor and then further by forecast quality, so that numerical runs with 50 microsecond baseline T1 reach approximately three to four times longer lifetimes together with improved population retention and integrated energy.

What carries the argument

The hybrid predictive feedback loop in which a fast-decaying ancilla supplies coherent feedback from both quadratures and a supervised predictor forecasts the homodyne current to compensate for latency.

If this is right

  • The ancilla suppresses the emission channel by a cooperativity factor.
  • The predictor further suppresses residual decay in proportion to forecast quality.
  • Numerical simulations with IBM-scale parameters achieve roughly three to four times longer T1 and better population retention.
  • The scheme is modular so that ancilla coupling and prediction can be added to existing Wiseman-Milburn loops.
  • Closed-form effective rates make the separate contributions of ancilla and predictor analytically transparent.

Where Pith is reading between the lines

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

  • The modular design suggests the same ancilla-plus-predictor addition could be layered onto other feedback or error-mitigation protocols for cumulative gains.
  • Real-device tests on IBM hardware would directly check whether the simulated lifetime gains survive calibration imperfections and measurement noise.
  • The predictor component might generalize to other delayed-control settings in open quantum systems where future measurement statistics can be learned.

Load-bearing premise

The ancilla can be made to decay much faster than the system while remaining coherently coupled without adding new decoherence channels, and the predictor can deliver accurate real-time forecasts of the homodyne current under realistic hardware latency.

What would settle it

A simulation or experiment in which the ancilla decay is not substantially faster than the system or the predictor accuracy is low, so that the measured effective T1 fails to exceed the standard Wiseman-Milburn result.

Figures

Figures reproduced from arXiv: 2511.13774 by Ali Abu-Nada, Aryan Iliat, Russell Ceballos.

Figure 1
Figure 1. Figure 1: Schematic of the feedback–controlled homodyne interferometer used [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Hybrid predictive feedback scheme. The emitted light from the system is measured by two detectors (D1, D2), producing the homodyne current I(t) = i1(t) − i2(t). A supervised machine-learning model predicts the future signal Ib(t+τ) to overcome the feedback delay τ. The predicted current drives the modulator, which adjusts the laser field to α + λIb(t+τ). This corrected field interacts with a coherently cou… view at source ↗
Figure 3
Figure 3. Figure 3: Digitizing the homodyne signal and forming training samples. The orange curve shows the measured homodyne current I(t) after analog￾to-digital conversion. Each black cross marks one of the five most recent samples [I(ti−5), . . . , I(ti−1)] that form the input vector xi, while the red point represents the next sample I(ti), which the network learns to predict. This sliding-window process converts the conti… view at source ↗
Figure 4
Figure 4. Figure 4: Prediction versus measured delayed current. The red curve shows the measured delayed current I(t+τ), while the blue curve shows the ML prediction Ib(t+τ) obtained from the input windows in Table I. The dashed gray curve indicates the present current I(t). The close overlap between red and blue confirms that the trained network accurately anticipates the future signal, allowing the feedback to stay synchron… view at source ↗
Figure 5
Figure 5. Figure 5: Neural network used for homodyne-current prediction. The input layer receives the last W=5 delayed samples [I(ti−5+τ), . . . , I(ti−1+τ)]. Two hidden layers (32 and 16 neurons) with ReLU activations capture nonlinear dependencies in the time series, and a linear output neuron provides the forecast Ib(ti+τ) used by the controller. timescale, the unconditional evolution of the atom’s density matrix ρ(t) is g… view at source ↗
Figure 8
Figure 8. Figure 8: plots the energy-retention curve, defined as the area under the excited-state population up to time T. Operationally, it is the cumulative time the qubit spends excited between 0 and T. Because the qubit’s internal energy is proportional to its excited-state population, this area is directly proportional to the total stored energy over the interval. Equivalently, it quantifies how much excited-state “budge… view at source ↗
Figure 7
Figure 7. Figure 7: Time evolution of the excited-state population [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
read the original abstract

Amplitude damping fundamentally limits qubit lifetimes by irreversibly leaking energy and information into the environment. Standard Wiseman--Milburn feedback offers only modest improvement because it acts on a single measured quadrature and its corrective drive is degraded by loop delay. We introduce a compact hybrid upgrade with two components: (i) a coherently coupled \emph{ancilla} qubit that receives the homodyne current and feeds back \emph{quantum-coherently} on the system, recovering information from \emph{both} field quadratures and intentionally engineered to decay much faster than the system; and (ii) a lightweight supervised predictor that forecasts the near-future homodyne current, phase-aligning the correction to overcome hardware latency. A Lindblad treatment yields closed-form effective decay rates: the ancilla suppresses the emission channel by a cooperativity factor, while the predictor further suppresses the residual decay in proportion to forecast quality. Using IBM-scale parameters (baseline \(T_1 = 50~\mu\mathrm{s}\)), numerical simulations surpass the W--M limit, achieving \(\sim 3\!-\!4\times\) longer \(T_1\) together with improved population retention and integrated energy. The method is modular and hardware-compatible: ancilla coupling and supervised prediction can be added to existing W--M loops to convert leaked information into a precise, time-advanced corrective drive. We also include a detailed, student-friendly derivation of the effective rates for both ancilla-assisted and prediction-enhanced feedback, making the impact of each design element analytically transparent.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript proposes a hybrid quantum feedback scheme for extending qubit lifetimes beyond the Wiseman-Milburn limit. It combines a coherently coupled fast-decaying ancilla qubit (engineered with Γ_ancilla ≫ Γ_system) that recovers information from both field quadratures with a supervised predictor that forecasts the homodyne current to compensate for hardware latency. A Lindblad master-equation treatment is used to derive closed-form effective decay rates, with the ancilla suppressing the emission channel by a cooperativity factor C and the predictor further reducing residual decay in proportion to forecast quality. Numerical simulations with IBM-scale parameters (baseline T1 = 50 μs) are reported to achieve ∼3–4× longer effective T1, together with improved population retention and integrated energy. The scheme is presented as modular and hardware-compatible, with detailed student-friendly derivations included.

Significance. If the central claims hold under realistic conditions, the work could be significant for quantum information processing by offering a modular upgrade to existing Wiseman-Milburn loops that converts leaked information into a time-advanced corrective drive. The inclusion of closed-form derivations and a student-friendly treatment of the effective rates is a clear strength, making the contribution of each design element analytically transparent. The numerical demonstrations with practical parameters add potential impact, although the result's dependence on idealized assumptions about ancilla engineering and predictor performance under latency limits the assessed significance at present.

major comments (2)
  1. [§3] §3 (Lindblad derivation of effective rates): The closed-form suppression of the emission channel by cooperativity factor C is derived under the assumption of perfect coherent coupling with no additional Lindblad terms for dephasing or back-action from the fast ancilla. This assumption is load-bearing for the analytical result and the claim of surpassing the W-M limit; the manuscript provides no quantitative robustness analysis or parameter scan showing how small violations (e.g., extra dephasing rates comparable to realistic hardware values) modify the effective decay formulas.
  2. [§5] §5 (numerical simulations and predictor): The reported ∼3–4× T1 improvement relies on the supervised predictor delivering high forecast quality that proportionally suppresses residual decay. The manuscript does not demonstrate that this forecast quality is obtained from independent measurement or derivation rather than fitting; per the paper's own equations relating suppression to forecast accuracy, this risks reducing the enhancement to a fitted parameter, undermining the predictive aspect of the hybrid scheme.
minor comments (2)
  1. [Notation] The definition and explicit formula for the cooperativity factor C should be stated in the main text (rather than only in the appendix) to improve readability of the effective-rate expressions.
  2. [Figures] Figure captions for the simulation results should explicitly list the IBM-scale parameters (including latency values and forecast-error metrics) used, to facilitate reproducibility.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed comments. We address each major comment point by point below, indicating revisions where the concerns identify areas for improvement.

read point-by-point responses

  1. Referee: [§3] §3 (Lindblad derivation of effective rates): The closed-form suppression of the emission channel by cooperativity factor C is derived under the assumption of perfect coherent coupling with no additional Lindblad terms for dephasing or back-action from the fast ancilla. This assumption is load-bearing for the analytical result and the claim of surpassing the W-M limit; the manuscript provides no quantitative robustness analysis or parameter scan showing how small violations (e.g., extra dephasing rates comparable to realistic hardware values) modify the effective decay formulas.

    Authors: We agree that the closed-form derivation in §3 relies on the ideal coherent coupling assumption without extra dephasing or back-action terms. This is a standard simplification to obtain transparent analytical expressions, but we recognize the need for robustness checks under realistic conditions. In the revised manuscript we will add a new subsection with a parameter scan that introduces additional dephasing rates on both system and ancilla (scaled to typical hardware values) and quantifies their impact on the effective decay rates and the cooperativity suppression factor. revision: yes

  2. Referee: [§5] §5 (numerical simulations and predictor): The reported ∼3–4× T1 improvement relies on the supervised predictor delivering high forecast quality that proportionally suppresses residual decay. The manuscript does not demonstrate that this forecast quality is obtained from independent measurement or derivation rather than fitting; per the paper's own equations relating suppression to forecast accuracy, this risks reducing the enhancement to a fitted parameter, undermining the predictive aspect of the hybrid scheme.

    Authors: The predictor coefficients are obtained from the known Lindblad dynamics and the homodyne current model rather than arbitrary fitting. To make this explicit and address the concern, the revised §5 will include a step-by-step derivation of the forecast from the master equation together with validation on held-out simulation trajectories. This will show that forecast quality follows from the model parameters and is not reduced to a post-hoc fitted quantity. revision: partial

Circularity Check

1 steps flagged

Effective decay rate reduction proportional to supervised forecast quality reduces to fitted input

specific steps
  1. fitted input called prediction [Abstract; Lindblad effective-rate derivation]
    "A Lindblad treatment yields closed-form effective decay rates: the ancilla suppresses the emission channel by a cooperativity factor, while the predictor further suppresses the residual decay in proportion to forecast quality. Using IBM-scale parameters (baseline T1 = 50 μs), numerical simulations surpass the W–M limit, achieving ∼3–4× longer T1"

    The residual decay term is scaled by forecast quality. Because the predictor is supervised (trained on homodyne data) and the reported lifetime gain is obtained from simulations that embed this quality factor, the suppression is a direct function of the fitted/trained accuracy rather than an independent prediction; the 'surpassing the limit' result is therefore statistically forced by the input parameter choice.

full rationale

The paper derives closed-form effective rates via Lindblad master equation where ancilla cooperativity C suppresses emission and predictor accuracy further scales the residual rate. However, the predictor is a supervised model whose forecast quality directly enters the rate formula; numerical simulations then report 3-4x T1 gain using this same quality factor. No independent derivation or external measurement of forecast quality is shown separate from the simulation outcomes, so the claimed extension beyond Wiseman-Milburn is forced by the input accuracy parameter rather than emerging from first principles. The ancilla engineering assumptions remain external but the predictor step matches the fitted-input-called-prediction pattern.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 1 invented entities

Central claim rests on the Lindblad master-equation framework for open quantum systems, the assumption that a fast ancilla can be coherently coupled without extra channels, and the existence of an accurate real-time predictor.

free parameters (2)
  • forecast quality
    Proportion of residual-decay suppression depends on this quantity; abstract does not state whether it is measured independently or fitted.
  • cooperativity factor
    Factor by which ancilla suppresses the emission channel; appears as a derived but tunable parameter.
axioms (2)
  • standard math Markovian approximation underlying the Lindblad equation
    Invoked for the derivation of effective decay rates.
  • domain assumption Ancilla decays much faster than the system qubit
    Explicitly stated as intentional engineering choice.
invented entities (1)
  • supervised predictor for homodyne current no independent evidence
    purpose: To forecast near-future signal and phase-align the correction against loop delay
    Introduced to overcome hardware latency; no independent evidence supplied in abstract.

pith-pipeline@v0.9.0 · 5580 in / 1548 out tokens · 52538 ms · 2026-05-17T21:54:59.539240+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

36 extracted references · 36 canonical work pages · 1 internal anchor

  1. [1]

    Non- markovian homodyne-mediated feedback on a two-level atom: a quantum trajectory treatment,

    J. Wang, H. Wiseman, and G. Milburn, “Non- markovian homodyne-mediated feedback on a two-level atom: a quantum trajectory treatment,”Chemical Physics, vol. 268, no. 1, pp. 221–235, 2001. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0301010401003044

    work page 2001

  2. [2]

    Breuer and F

    H.-P. Breuer and F. Petruccione,The Theory of Open Quantum Systems. Oxford University Press, 2002

    work page 2002

  3. [3]

    M. A. Nielsen and I. L. Chuang,Quantum computation and quantum information. Cambridge University Press, 2010

    work page 2010

  4. [4]

    A quantum engineer’s guide to superconducting qubits,

    P. Krantz, M. Kjaergaard, F. Yan, T. P. Orlando, S. Gustavsson, and W. D. Oliver, “A quantum engineer’s guide to superconducting qubits,” Applied Physics Reviews, vol. 6, no. 2, p. 021318, 2019

    work page 2019

  5. [5]

    Optimizing quantum gates towards the scale of logical qubits,

    P. V . Klimovet al., “Optimizing quantum gates towards the scale of logical qubits,”Nature Communications, vol. 15, p. 3993, 2024

    work page 2024

  6. [6]

    Surface codes: Towards practical large-scale quantum computation,

    A. G. Fowler, M. Mariantoni, J. M. Martinis, and A. N. Cleland, “Surface codes: Towards practical large-scale quantum computation,”Physical Review A, vol. 86, no. 3, p. 032324, 2012

    work page 2012

  7. [7]

    Spontaneous emission probabilities at radio frequencies,

    E. M. Purcell, “Spontaneous emission probabilities at radio frequencies,” Physical Review, vol. 69, p. 681, 1946

    work page 1946

  8. [8]

    Reaching 10 ms single photon lifetimes for superconducting aluminum cavities,

    M. Reagor, H. Paik, G. Catelani, L. Sun, C. Axline, E. T. Holland, I. M. Pop, N. A. Masluk, T. Brecht, L. Frunzio, M. H. Devoret, and R. J. Schoelkopf, “Reaching 10 ms single photon lifetimes for superconducting aluminum cavities,”Applied Physics Letters, vol. 102, no. 19, p. 192604, 2013

    work page 2013

  9. [9]

    Rapid high- fidelity multiplexed readout of superconducting qubits,

    J. Heinsoo, C. K. Andersen, A. Remm, S. Krinner, T. Walter, Y . Salath ´e, S. Gasparinetti, J.-C. Besse, A. Wallraff, and C. Eichler, “Rapid high- fidelity multiplexed readout of superconducting qubits,”Physical Review Applied, vol. 10, no. 3, p. 034040, 2018

    work page 2018

  10. [10]

    Fast accurate state measurement with superconducting qubits,

    E. Jeffrey, D. Sank, J. Y . Mutuset al., “Fast accurate state measurement with superconducting qubits,”Physical Review Letters, vol. 112, p. 190504, 2014

    work page 2014

  11. [11]

    Quantum theory of a bandpass purcell filter for qubit readout,

    E. A. Sete, J. M. Martinis, and A. N. Korotkov, “Quantum theory of a bandpass purcell filter for qubit readout,”Physical Review A, vol. 92, p. 012325, 2015

    work page 2015

  12. [12]

    Fast multiplexed superconducting-qubit readout with intrinsic tunable purcell filtering,

    P. A. Springet al., “Fast multiplexed superconducting-qubit readout with intrinsic tunable purcell filtering,”PRX Quantum, vol. 6, p. 020345, 2025

    work page 2025

  13. [13]

    Inhibited spontaneous emission in solid-state physics and electronics,

    E. Yablonovitch, “Inhibited spontaneous emission in solid-state physics and electronics,”Physical Review Letters, vol. 58, no. 20, pp. 2059– 2062, 1987

    work page 2059

  14. [14]

    Interfacing single photons and single quantum dots with photonic nanostructures,

    P. Lodahl, S. Mahmoodian, and S. Stobbe, “Interfacing single photons and single quantum dots with photonic nanostructures,”Reviews of Modern Physics, vol. 87, no. 2, pp. 347–400, 2015

    work page 2015

  15. [15]

    How to enhance dephasing time in superconducting qubits,

    Ł. Cywi ´nski, R. M. Lutchyn, C. P. Nave, and S. Das Sarma, “How to enhance dephasing time in superconducting qubits,”Physical Review B, vol. 77, no. 17, p. 174509, May 2008

    work page 2008

  16. [16]

    Optimized dynamical decoupling in a model quantum memory,

    M. J. Biercuk, H. Uys, A. P. VanDevender, N. Shiga, W. M. Itano, and J. J. Bollinger, “Optimized dynamical decoupling in a model quantum memory,”Nature, vol. 458, pp. 996–1000, 2009

    work page 2009

  17. [17]

    Optimized noise filtration through dynamical decoupling,

    H. Uys, M. J. Biercuk, and J. J. Bollinger, “Optimized noise filtration through dynamical decoupling,”Physical Review Letters, vol. 103, p. 040501, 2009

    work page 2009

  18. [18]

    Noise spectroscopy through dynamical decoupling with a superconducting flux qubit,

    J. Bylander, S. Gustavsson, F. Yan, F. Yoshihara, K. Harrabi, G. Fitch, D. G. Cory, Y . Nakamura, J.-S. Tsai, and W. D. Oliver, “Noise spectroscopy through dynamical decoupling with a superconducting flux qubit,”Nature Physics, vol. 7, no. 7, pp. 565–570, Jul. 2011

    work page 2011

  19. [19]

    Dynamics and control of two coupled quantum oscillators: An analytical approach,

    A. Abu-Nada and L.-A. Wu, “Dynamics and control of two coupled quantum oscillators: An analytical approach,” 2025. [Online]. Available: https://arxiv.org/abs/2509.21492

    work page arXiv 2025

  20. [20]

    Dynamical decoupling of open quantum systems,

    L. Viola and S. Lloyd, “Dynamical decoupling of open quantum systems,”Physical Review Letters, vol. 82, pp. 2417–2421, 1999

    work page 1999

  21. [21]

    Decoherence-free subspaces and subsystems,

    D. A. Lidar, L.-A. Wu, and P. Zanardi, “Decoherence-free subspaces and subsystems,”Journal of Modern Optics, vol. 50, no. 6-7, pp. 1047–1057, 2003

    work page 2003

  22. [22]

    Universal dynamical decoupling and quantum error suppression,

    M. S. Byrd and D. A. Lidar, “Universal dynamical decoupling and quantum error suppression,”Physical Review A, vol. 67, p. 012324, 2003

    work page 2003

  23. [23]

    Efficient dynamical decoupling of multiqubit states,

    L.-A. Wu, M. S. Byrd, and D. A. Lidar, “Efficient dynamical decoupling of multiqubit states,”Physical Review Letters, vol. 89, no. 12, p. 127901, 2002

    work page 2002

  24. [24]

    Dy- namical decoupling for superconducting qubits: A performance survey,

    N. Ezzell, B. Pokharel, L. Tewala, G. Quiroz, and D. A. Lidar, “Dy- namical decoupling for superconducting qubits: A performance survey,” Physical Review Applied, vol. 20, no. 6, p. 064027, Dec. 2023

    work page 2023

  25. [25]

    Introduction to quantum noise, measurement, and amplifi- cation,

    A. A. Clerk, M. H. Devoret, S. M. Girvin, F. Marquardt, and R. J. Schoelkopf, “Introduction to quantum noise, measurement, and amplifi- cation,”Reviews of Modern Physics, vol. 82, no. 2, pp. 1155–1208, Jun. 2010

    work page 2010

  26. [26]

    De- coherence in a superconducting quantum bit circuit,

    G. Ithier, E. Collin, P. Joyez, P. J. Meeson, D. Vion, D. Esteve, F. Chiarello, A. Shnirman, Y . Makhlin, J. Schriefl, and G. Sch ¨on, “De- coherence in a superconducting quantum bit circuit,”Physical Review B, vol. 72, no. 13, p. 134519, Oct. 2005

    work page 2005

  27. [27]

    Dynamical decoupling sequence construction as a filter-design problem,

    T. J. Green, J. Sastrawan, H. Uys, and M. J. Biercuk, “Dynamical decoupling sequence construction as a filter-design problem,”New Journal of Physics, vol. 15, p. 095004, 2013

    work page 2013

  28. [28]

    Quantum theory of optical feedback via homodyne detection,

    H. M. Wiseman and G. J. Milburn, “Quantum theory of optical feedback via homodyne detection,”Physical Review Letters, vol. 70, no. 5, pp. 548–551, 1993

    work page 1993

  29. [29]

    Cambridge University Press, 2009

    ——,Quantum Measurement and Control. Cambridge University Press, 2009

    work page 2009

  30. [30]

    Quantum feedback control and classical control theory,

    A. C. Doherty and K. Jacobs, “Quantum feedback control and classical control theory,”Physical Review A, vol. 60, no. 4, pp. 2700–2711, 1999

    work page 1999

  31. [31]

    H. J. Carmichael,An Open Systems Approach to Quantum Optics, ser. Lecture Notes in Physics Monographs. Springer, 1993

    work page 1993

  32. [32]

    A straightforward introduction to continuous quantum measurement,

    K. Jacobs and D. A. Steck, “A straightforward introduction to continuous quantum measurement,”Contemporary Physics, vol. 47, no. 5, pp. 279– 303, 2006

    work page 2006

  33. [33]

    Supervised machine learning for predicting open quantum system dynamics and detecting non-markovian memory effects,

    A. Abu-Nada and S. Banerjee, “Supervised machine learning for predicting open quantum system dynamics and detecting non-markovian memory effects,” 2025. [Online]. Available: https://arxiv.org/abs/2509.22758

    work page arXiv 2025

  34. [34]

    Rectified linear units improve restricted boltz- mann machines,

    V . Nair and G. E. Hinton, “Rectified linear units improve restricted boltz- mann machines,” inProceedings of the 27th International Conference on Machine Learning (ICML-10), 2010, pp. 807–814

    work page 2010

  35. [35]

    Adam: A Method for Stochastic Optimization

    D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” arXiv preprint arXiv:1412.6980, 2014. TABLE II COMPARISON OF FEEDBACK SCHEMES FOR AMPLITUDE-DAMPING SUPPRESSION. HEREP e(t)IS THE EXCITED-STATE POPULATION,Γ effTHE EFFECTIVE DECAY RATE,ANDT eff 1 = 1/ΓeffTHE CORRESPONDING LIFETIME. SchemeP e(t)Γ eff Teff 1 No feedbackP e(0)e −γt γ1/γ W–...

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  36. [36]

    Quantum feedback networks: modelling and synthesis,

    J. E. Gough, “Quantum feedback networks: modelling and synthesis,” Proceedings of the Royal Society A, vol. 468, no. 2146, p. 2189–2204, 2010. APPENDIXA DERIVATION OFEQ. (17) Starting from the master equation with spontaneous emis- sion at rateγ, ˙ρ(t) =γD[σ−]ρ(t) =γ σ−ρσ+ − 1 2 {σ+σ−, ρ} ,(31) where the Lindblad dissipator is defined as D[L]ρ=LρL † − 1 2...

    work page 2010