ResQ: A Novel Framework to Implement Residual Neural Networks on Analog Rydberg Atom Quantum Computers

Jason Han; Nicholas S. DiBrita; Tirthak Patel

REVIEW 4 major objections 4 minor 40 references

ResQ: A Novel Framework to Implement Residual Neural Networks on Analog Rydberg Atom Quantum Computers

T0 review · 4 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read RESQ maps residual networks onto Rydberg atoms and claims a 56% edge over classical peers.

desk verdict Useful Rydberg pulse-encoding scheme, but the quantum-ResNet framing and the outperformance claim don't survive close reading. read the letter →

arxiv 2506.21537 v1 pith:GSGZZA4N submitted 2025-06-26 quant-ph cs.CVcs.ET

Nicholas S. DiBrita , Jason Han , Tirthak Patel This is my paper

classification quant-phcs.CVcs.ET MSC 81P6868T05 PACS 03.67.Ac03.67.Lx

keywords quantummachinelearningRydbergatomcomputerresidualneuralnetworkordinarydifferentialequationsanalogcomputingHamiltonianparameterizationMNISTclassificationnoiserobustness

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

RESQ introduces a framework for implementing residual neural networks on analog Rydberg-atom quantum computers by encoding input features and trainable parameters directly into the Hamiltonian pulses that drive the atoms. The paper claims this is the first fully quantum-native realization of ResNets, and reports that on binary classification tasks from MNIST, FashionMNIST, and a diabetes dataset it outperforms similarly sized classical feedforward, residual, and neural-ODE networks by 56%, 57%, and 36%, respectively, while remaining robust to hardware noise on QuEra's Aquila device. The motivation is that ResNets have a continuous-time ODE structure that matches the natural Schrödinger dynamics of analog quantum hardware, whereas gate-based quantum computers can only approximate such dynamics with many discrete steps. If the claims hold, RESQ offers a parameter-efficient recipe for building residual architectures natively on near-term analog quantum machines.

What carries the argument

The load-bearing object is the parameterized analog Hamiltonian $$H(t)=\frac{\$\Omega$(t)}{2}\sum_i(e^{i\$\varphi$(t)}|g\rangle_i\langle r|_i+e^{-i\$\varphi$(t)}|r\rangle_i\langle g|_i)-\$\Delta$(t)\sum_i \hat n_i+\sum_{i<j}\frac{C_6}{|\vec p_i-\vec p_j|^6}\hat n_i\hat n_j-\delta(t)\sum_i h_i \hat n_i,$$ with $\Omega(t)$, $\Delta(t)$, and $\delta(t)$ specified as piecewise-linear pulses. RESQ's central parameterization sets each pulse strength at a holding time to $\theta_j \omega_i+\theta_{j+1}$, so a single input feature $\omega_i$ is scaled and offset by learned parameters throughout the evolution, while alternating local couplings $h_i$ lift further features and parameters into the computation. This turns the hardware's continuous-time evolution into the residual/neural-ODE structure: the input is not fed through fixed layers but shapes the entire state trajectory, and the final averaged $\langle|1\rangle$ probability is the classifier output. The stochastic pulse-gradient method supplies unbiased gradients of this analog program with respect to the Hamiltonian parameters.

What would settle it

Train the same classical feedforward, residual, and neural-ODE classifiers on the same five PCA features but let them run to convergence (for example, 1,000 or 10,000 Adam iterations instead of 75) and compare final test accuracy; if any reaches or exceeds RESQ's reported accuracy, the claimed 56%, 57%, and 36% improvements are artifacts of the training-time cutoff rather than a property of the quantum model.

Watch

Extended reading notes

Core claim

RESQ claims that an analog Rydberg-atom quantum computer can implement residual neural networks directly, without gates, by treating the Schrödinger evolution itself as a trainable neural ODE. The framework encodes each input feature $\omega_i$ into the pulse heights of the global Rabi frequency, global detuning, and local detuning, with each height written as $\theta_j\omega_i+\theta_{j+1}$ for learned parameters $\theta_j, \theta_{j+1}$, and encodes additional features and parameters into alternating site-dependent couplings $h_i$ of the local detuning term. After evolving the $N$-atom system for a fixed time, the averaged probability of measuring atoms in the $|1\rangle$ state is the soft label, and the parameters are trained by backpropagating cross-entropy loss through an unbiased stochastic pulse-gradient estimator. With $N=4$ atoms and three pulse intervals, this gives 20 trainable parameters and five input features; on binary MNIST, FashionMNIST, and diabetes tasks the paper reports 56%, 57%, and 36% accuracy improvements over similarly sized classical feedforward, residual, and neural-ODE networks, respectively, with accuracy and F1 within about 1% of ideal when pulse and position noise is simulated. Real-hardware inference on Aquila is reported to be largely robust except for samples near the decision boundary.

Load-bearing premise

The headline 'similarly sized classical models' comparison assumes that a classical network trained on the same five PCA features for the same 75 iterations is a meaningful baseline; if classical models are instead given raw data, more features, or longer training, RESQ's reported improvements are not established.

Editorial extensions

If this is right

RESQ gives current analog hardware a usable classification recipe: with four atoms it uses 20 trainable parameters and five input features, and inference on Aquila is reported to stay within about 1% of ideal accuracy under simulated noise.
Because features enter through pulse heights and local couplings, the number of input features grows linearly with qubit count ($3+N/2$ for $N$ atoms) while the evolution time stays fixed, so larger atom arrays could handle richer inputs without deeper circuits.
Residual structure is realized natively as continuous Hamiltonian evolution, which means analog Rydberg systems can bypass the discrete-step decomposition that gate-based quantum ResNet proposals require.
The framework is not tied to images: the same pipeline is demonstrated on a tabular healthcare dataset, suggesting generality across binary classification tasks.
If the results hold, RESQ is the first fully quantum-native neural-ODE classifier, and the mapping from model parameters to Hamiltonian pulses becomes a general template for other continuous-time models.

Reading between the lines

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

In our reading, the reported advantage is more plausibly a statement about baseline starvation than about quantum expressivity: classical architectures trained on the full images, or on the five PCA features with many more iterations, would likely close most of the gap. We would test this before treating the 56% figure as a quantum speedup.
The residual block analogy is conceptual, not literal: Schrödinger evolution is unitary, so RESQ implements a continuous-time integration of a linear operator rather than the additive nonlinear skip layer of the original ResNet; the natural frame for the result is analog quantum neural ODEs.
A direct scaling test would run RESQ with $N>4$ atoms on Aquila to see whether the linear feature-scaling claim survives real interaction graphs and noise; the paper's hardware evaluation is limited to the same $N=4$ configuration used in simulation.
Because the gradient estimator is unbiased and hardware-agnostic, the same piecewise-pulse parameterization could be transferred to other controllable analog platforms, so the framework's value may outlive Rydberg-specific hardware.

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.

Desk Editor's Note

read the letter

First: the encoding scheme is a real contribution. RESQ parameterizes pulse amplitudes as piecewise-linear functions θ_j ω_i + θ_{j+1}, uses local detuning couplings to embed features, and demonstrates this on QuEra Aquila with code and data available. The linear scaling of input features with qubit count is a nice property, and the authors deserve credit for testing on real hardware.

The problem is that the two headline claims are not supported. The paper nowhere derives a residual-network mapping. The Schrödinger evolution is linear in the state; the Trotterized update is |ψ(t+dt)⟩ ≈ (I − iH(t)dt)|ψ(t)⟩, with no nonlinear residual function F(x), no skip connection, and no state variable playing the role of a data representation. The only nonlinearity comes from the final Born-rule measurement. Calling this a quantum neural ODE is an analogy, not a derivation, and the residual claim is load-bearing for the paper's novelty.

The performance advantage is also an artifact of the baseline setup. Classical models receive the same 5 PCA features and only 75 full-batch iterations, with no hyperparameter tuning. Near-chance scores like 51% on MNIST 0/1 show these baselines are undertrained. So the reported 56%, 57%, and 36% improvements are not evidence of any quantum advantage. The hardware evaluation uses 8 samples per task without error bars, and the simulated-noise claims are given without error bars either.

What is left is a plausible engineering recipe for analog QML. Someone working on Rydberg classifiers may want to borrow the pulse parameterization and the local/global encoding split. But the paper oversells it as a first quantum ResNet. The authors should either supply a genuine derivation of the residual structure or drop that language, and they must redo the classical baseline comparison with proper tuning, multiple seeds, and error bars. The footnote stating that the paper is already published at ICCV 2025 also needs to be resolved if this is being submitted elsewhere.

This is a paper that deserves referee time—the engineering is concrete and reproducible—but my verdict would be major revision, not acceptance. I'd send it to reviewers with explicit instructions to check the residual mapping and the baseline fairness.

Referee Report

4 major / 4 minor

Summary. The paper introduces RESQ, a framework that encodes PCA-reduced input features and trainable parameters into the time-dependent pulse sequences of an analog Rydberg atom quantum computer. The dynamics are governed by the Schrödinger equation with a globally and locally addressed Hamiltonian (Eqs. (1)-(2)), and classification is performed by measuring the average excited-state probability. The authors claim that this constitutes the first implementation of residual neural networks (ResNets) and neural ODEs on analog Rydberg hardware, and that RESQ outperforms similarly sized classical feedforward networks, ResNets, and neural ODE classifiers by 56%, 57%, and 36%, respectively, while remaining robust to hardware noise. Evaluation is performed on MNIST, FashionMNIST, and the Pima Indians Diabetes dataset using classical simulation, with a small set of inference runs on the QuEra Aquila device.

Significance. If the central claims were valid, the paper would represent a notable advance in analog quantum machine learning, combining continuous Hamiltonian evolution with a residual-network-style architecture and demonstrating a practical advantage over classical models. The paper also has concrete strengths: the code and data are open-sourced, the authors use a real 256-qubit Rydberg device for inference, and they explore several lattice geometries and spacings. However, the significance is substantially weakened by two load-bearing problems: (i) the residual-network interpretation is not formally justified, since the Hamiltonian evolution is linear and state-independent, and (ii) the classical baselines are handicapped by receiving the same 5 PCA features and only 75 training iterations, making the reported improvements unsurprising. The hardware robustness claim rests on only 8 samples per task, which cannot support the stated conclusion. The result, as presented, does not establish a quantum advantage or a genuine implementation of residual networks.

major comments (4)

[§4.1, Eq. (1)] The central architectural claim is that RESQ implements residual neural networks, but the dynamics in Eq. (1) do not contain a residual function F(x, θ) of the form defined in §2.2. The Hamiltonian H(t) depends on the input features ω_i and on trainable parameters, but not on the instantaneous quantum state |ψ(t)⟩. Consequently, the Trotterized update |ψ(t+dt)⟩ ≈ (I - iH(t)dt)|ψ(t)⟩ is a linear, state-independent map; the only nonlinearity in the entire model is the final Born-rule measurement. Calling this an implementation of ResNets or neural ODEs is an analogy, not a derivation. This undermines the abstract's primary claim of being 'the first framework to implement residual neural networks on analog Rydberg atom quantum computers.'
[§5, 'Comparative Classical Techniques'] The performance comparison is not fair. The classical baselines C-NN, C-ResNet, and C-NODE receive the same 5 PCA features as RESQ and are trained for only 75 full-batch iterations with no minibatching and no hyperparameter tuning. Under these conditions, near-chance accuracies such as 51% on MNIST 0/1 (Figure 6) indicate severe undertraining, so the reported improvements of 56%, 57%, and 36% in Section 6.1 are artifacts of weak baselines rather than evidence of a quantum advantage. A meaningful comparison would allow classical models to use the raw data or a reasonable feature representation, and would use standard training budgets with hyperparameter selection.
[§6.4, 'Real Hardware Executions'] The robustness claim on real hardware is based on only 8 samples per classification task. With n=8, the observation that some points near the decision boundary flip under noise is anecdotal and cannot support the conclusion that 'RESQ has excellent resilience to noise.' The simulated-noise analysis reportedly shows accuracy within 1% of ideal, but no error bars, confidence intervals, or detailed statistics are provided. This is insufficient support for the paper's robustness claim.
[§4.4 and §6.1] The paper does not ablate the choice of 5 PCA features or compare against classical models using more features or raw pixel inputs. Since the classical baselines are restricted to the same 5 features, the comparison conflates the feature representation with the model architecture. Without such ablations, the statement that RESQ 'outperforms similarly sized classical models' is not meaningful, because the classical models are artificially limited in their input information.

minor comments (4)

[Throughout] The name 'RESQ' is typeset inconsistently as 'R ESQ' in several places; please use a consistent notation.
[Figure 5 caption] The caption says 'Pluses used by RESQ'; this should read 'Pulses used by RESQ.'
[§1] The statement that Rydberg atom systems 'are currently the only hardware that supports continuous-time Hamiltonian evolution and local/global control' is too strong and unsubstantiated; other quantum platforms (e.g., trapped ions) also offer continuous control, and the claim should be qualified or removed.
[§5, 'Software and Simulation Setup'] The training hyperparameters are underspecified. The paper mentions Adam and 20 gradient samples but does not report learning rate, number of gradient steps per iteration, or initialization details beyond 'all parameters initialized to 1.0,' which would be needed to reproduce the results.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the derivation is self-contained and the reported predictions are measured on held-out data, not reconstructed from fitted parameters.

full rationale

RESQ's derivation chain is self-contained: the Hamiltonian in Eq. (1) is the standard Rydberg Hamiltonian; inputs and trainable parameters enter only through the piecewise-linear pulse parameterization θ_j ω_i + θ_{j+1} (Sec. 4.3) and the local couplings h_i; and the prediction is the Born-rule average P(|1⟩) over all measured qubits on a held-out test set (Sec. 4.5). No parameter is fitted to the test labels, no test-set statistic is renamed as a prediction, and no uniqueness theorem from the authors is invoked to force the design. The claim that RESQ implements residual networks rests on the standard continuous-time equivalence between ResNets and neural ODEs (dx/dt = F(x,θ), Sec. 2.2) applied to the Schrödinger equation; this is an analogy that the paper asserts rather than derives at the equation level, but it is not a circular reduction of the result to its inputs. The only self-citation ([12]) supplies standard Rydberg background and a noise model; it is not load-bearing, and the gradient method and noise model are independently supported by external references [23] and [26]. Concerns about the weak classical baselines (same 5 PCA features, only 75 iterations, near-chance accuracies) are evaluation-quality issues, not circularity.

Assumptions & free parameters 8 free parameters · 6 assumptions · 0 invented entities

The central claim rests on the unproven equivalence between Rydberg Hamiltonian dynamics and residual networks, plus several hand-chosen hyperparameters (M=3, 5 PCA features, spacing, 20 gradient samples, 75 iterations). The trained pulse parameters are fit to data, so the reported accuracy reflects fitting, not prediction. The only standard input is Schrödinger's equation.

free parameters (8)

Pulse and local coupling parameters (total 6M + N/2) = 20 for M=3, N=4
Trained by gradient descent on 1000-sample training sets; these are the model weights.
Number of PCA features = 5
Chosen by hand; high-variance features assigned to global pulses for best reported performance.
Pulse interval count M = 3
Selected after ablation in Section 6.3, balancing accuracy and simulation time.
Lattice spacing = 12 micrometers or similar moderate spacing
Selected after ablation in Section 6.2; moderate weak-interaction spacing preferred.
Lattice configuration = square or chain for most tasks
Selected per dataset after ablation; ring, square, and triangle explored.
Gradient samples per update = 20
Empirical balance between accuracy and simulation time, Section 4.6.
Training iterations = 75
Fixed for all models; likely too few for classical baselines to converge.
Input scaling ranges = Omega and Delta in [pi/2, 2pi], h_i in [0,1]
Chosen to satisfy hardware constraints and force non-trivial dynamics, Section 4.4.

assumptions (6)

domain assumption The Rydberg Hamiltonian (Eq. 1) with global and local detuning accurately describes the Aquila hardware and its noise.
Used throughout Sections 4 and 6.4; noise modeled as Gaussian perturbations to pulses and positions.
ad hoc to paper Continuous Schrödinger evolution under a time-dependent Hamiltonian is equivalent to a residual neural network or neural ODE.
Asserted in Sections 2.3 and 4 without derivation; the Hamiltonian is linear in the state while neural ODEs use nonlinear F.
domain assumption Averaging the measured |1> probabilities over all qubits yields a meaningful scalar class score for binary classification.
Introduced in Section 4.5; no analysis of why averaging preserves class information.
ad hoc to paper The piecewise-linear pulse parameterization theta_j omega_i + theta_{j+1} has sufficient expressivity for the classification tasks.
Adopted in Section 4.3; no expressivity analysis; M=3 fixed after ablation.
domain assumption PCA features with MinMax scaling preserve class-discriminative information for the quantum encoding.
Used in Section 4.4; classical baselines may be handicapped if they receive the same reduced features.
standard math Schrödinger equation i hbar d|psi>/dt = H(t)|psi> governs the atom array.
Used in Sections 2.3 and 4.1 as the basis of analog evolution.

how reviews work

0 comments

Cite this review

Pith. "Pith review of ResQ: A Novel Framework to Implement Residual Neural Networks on Analog Rydberg Atom Quantum Computers." pith.science (2026). https://pith.science/paper/GSGZZA4N

@misc{pith2026250621537,
  author       = {Pith},
  title        = {Pith review of: ResQ: A Novel Framework to Implement Residual Neural Networks on Analog Rydberg Atom Quantum Computers},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GSGZZA4N}},
  note         = {Machine review of arXiv:2506.21537}
}

read the original abstract

Research in quantum machine learning has recently proliferated due to the potential of quantum computing to accelerate machine learning. An area of machine learning that has not yet been explored is neural ordinary differential equation (neural ODE) based residual neural networks (ResNets), which aim to improve the effectiveness of neural networks using the principles of ordinary differential equations. In this work, we present our insights about why analog Rydberg atom quantum computers are especially well-suited for ResNets. We also introduce ResQ, a novel framework to optimize the dynamics of Rydberg atom quantum computers to solve classification problems in machine learning using analog quantum neural ODEs.

Figures

Figures reproduced from arXiv: 2506.21537 by the authors.

**Figure 1.** High-level representation of RESQ’s analog quantum implementation of the classical residual neural network (ResNet). These residual blocks can help circumvent issues that arise in training via gradient-based methods [16]. Most near-term QML architectures do not have the ability to perform this skip operation, as quantum gates only allow for unitary linear transformations of the model’s state. To overcome this, we b… view at source ↗

**Figure 2.** Representation of RESQ’s classification workflow for the example of classifying 4 vs. 9 in the MNIST dataset [11]. Note that RESQ uses the measurement of all the qubits. with analog quantum computers, unlike digital quantum and classical architectures. Rydberg atom systems, in particular, are currently the only hardware that supports continuoustime Hamiltonian evolution and local/global control, making them unique… view at source ↗

**Figure 3.** (a) RESQ explores different atom configuration grids: chain, ring, square, and triangle. (b) RESQ explores different grid scales. without decomposing them into many small discrete steps. In contrast, analog quantum systems (such as Rydberg atom arrays) offer native support for continuous time evolution, making them especially well-suited for implementing such models directly in hardware. 2.2. Residual Neural Network… view at source ↗

Figures from the paper (7 more)

**Figure 4.** Figure 4: RESQ divides the pulses into multiple intervals (five in this example), with two parameters per interval, including the scaling for the input feature (ωi) and the offset. lution, the qubits are measured, and the probability of measuring |1⟩ (averaged across all qubits…

**Figure 5.** Figure 5: Pluses used by RESQ. All local detuning pulses (δi) have the same pulse shape, which is individually scaled for each qubit – per the hi value. Even qubit hi’s encode the input features, and odd qubit hi’s encode the parameters. tion for Ω, ∆, and δ, giving 3 inputs to …

**Figure 6.** Figure 6: Accuracy scores for the Pima Indian Diabetes dataset [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]

**Figure 7.** Figure 7: F1 scores for MNIST classification tasks (top) and Fash [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]

**Figure 9.** Figure 9: Accuracy on the Pima Indian Diabetes dataset using the [PITH_FULL_IMAGE:figures/full_fig_p008_9.png]

**Figure 8.** Figure 8: Performance of RESQ for different atom configurations and spacings as tested on different pairs of MNIST classes. In the FashionMNIST tasks, we see a smaller but still substantial 37% improvement over the various small classical models. In the trouser/boot classificat…

**Figure 10.** Figure 10: Prediction examples for MNIST and FashionMNIST datasets, using both ideal simulation and a real Rydberg atom computer. [PITH_FULL_IMAGE:figures/full_fig_p009_10.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

40 extracted references · 36 canonical work pages

[1]

https://www.kaggle.com/datasets/ uciml/pima-indians-diabetes-database, 2024

Pima Indians Diabetes Database. https://www.kaggle.com/datasets/ uciml/pima-indians-diabetes-database, 2024. Accessed: 2024-10-01. 7

work page 2024
[2]

Quantum computing optimization technique for iot platform using modified deep residual approach

Rasha M Abd El-Aziz, Ahmed I Taloba, and Fahad A Al- ghamdi. Quantum computing optimization technique for iot platform using modified deep residual approach. Alexandria Engineering Journal, 61(12):12497–12509, 2022. 4

work page 2022
[3]

AWS Aquila Interface

Amazon-Braket. AWS Aquila Interface. https : / / github . com / amazon - braket / amazon - braket - examples / blob / main / examples / analog _ hamiltonian _ simulation / 01 _ Introduction _ to _ Aquila . ipynb, 2024. Ac- cessed: 2024-11-01. 7

work page 2024
[4]

Qure: Qubit re-allocation in noisy intermediate-scale quan- tum computers

Abdullah Ash-Saki, Mahabubul Alam, and Swaroop Ghosh. Qure: Qubit re-allocation in noisy intermediate-scale quan- tum computers. In Proceedings of the 56th Annual Design Automation Conference (DAC), pages 1–6, 2019. 2

work page 2019
[5]

Measuring analytic gradients of general quantum evolution with the stochastic parameter shift rule

Leonardo Banchi and Gavin E Crooks. Measuring analytic gradients of general quantum evolution with the stochastic parameter shift rule. Quantum, 5:386, 2021. 4, 6

work page 2021
[6]

Muqut: Multi-constraint quantum circuit mapping on nisq computers

Debjyoti Bhattacharjee, Abdullah Ash Saki, Mahabubul Alam, Anupam Chattopadhyay, and Swaroop Ghosh. Muqut: Multi-constraint quantum circuit mapping on nisq computers. In 2019 IEEE/ACM international conference on computer-aided design (ICCAD), pages 1–7. IEEE, 2019. 2

work page 2019
[7]

Quantum ma- chine learning

Jacob Biamonte, Peter Wittek, Nicola Pancotti, Patrick Rebentrost, Nathan Wiebe, and Seth Lloyd. Quantum ma- chine learning. Nature, 549(7671):195–202, 2017. 1

work page 2017
[8]

Challenges and opportu- nities in quantum machine learning

Marco Cerezo, Guillaume Verdon, Hsin-Yuan Huang, Lukasz Cincio, and Patrick J Coles. Challenges and opportu- nities in quantum machine learning. Nature Computational Science, 2(9):567–576, 2022. 1

work page 2022

Show all 40 references

[9]

Neural ordinary differential equa- tions

Ricky TQ Chen, Yulia Rubanova, Jesse Bettencourt, and David K Duvenaud. Neural ordinary differential equa- tions. Advances in neural information processing systems , 31, 2018. 1, 3

2018
[10]

Learning quantum dynamics with la- tent neural ordinary differential equations.Phys

Matthew Choi, Daniel Flam-Shepherd, Thi Ha Kyaw, and Al´an Aspuru-Guzik. Learning quantum dynamics with la- tent neural ordinary differential equations.Phys. Rev. A, 105: 042403, 2022. 4

2022
[11]

The MNIST Database of Handwritten Digit Images for Machine Learning Research

Li Deng. The MNIST Database of Handwritten Digit Images for Machine Learning Research. IEEE Signal Processing Magazine, 29(6):141–142, 2012. 2, 7

2012
[12]

DiBrita, Daniel Leeds, Yuqian Huo, Jason Lud- mir, and Tirthak Patel

Nicholas S. DiBrita, Daniel Leeds, Yuqian Huo, Jason Lud- mir, and Tirthak Patel. ReCon: Reconfiguring Analog Ry- dberg Atom Quantum Computers for Quantum Generative Adversarial Networks. In Proceedings of the 43rd Inter- national Conference on Computer-Aided Design (ICCAD) ,
[13]

Quantum reinforcement learning

Daoyi Dong, Chunlin Chen, Hanxiong Li, and Tzyh-Jong Tarn. Quantum reinforcement learning. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 38 (5):1207–1220, 2008. 1

2008
[14]

Aug- mented neural odes

Emilien Dupont, Arnaud Doucet, and Yee Whye Teh. Aug- mented neural odes. Advances in neural information pro- cessing systems, 32, 2019. 7

2019
[15]

Deep residual learning in spiking neural networks

Wei Fang, Zhaofei Yu, Yanqi Chen, Tiejun Huang, Timoth´ee Masquelier, and Yonghong Tian. Deep residual learning in spiking neural networks. Advances in Neural Information Processing Systems, 34:21056–21069, 2021. 3

2021
[16]

Deep residual learning for image recognition

Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceed- ings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016. 1, 3

2016
[17]

Power of data in quantum machine learning

Hsin-Yuan Huang, Michael Broughton, Masoud Mohseni, Ryan Babbush, Sergio Boixo, Hartmut Neven, and Jarrod R McClean. Power of data in quantum machine learning. Na- ture communications, 12(1):2631, 2021. 1

2021
[18]

Learning to predict arbitrary quantum processes

Hsin-Yuan Huang, Sitan Chen, and John Preskill. Learning to predict arbitrary quantum processes. PRX Quantum, 4(4): 040337, 2023. 1

2023
[19]

Resqnets: a residual approach for mitigating barren plateaus in quantum neural networks

Muhammad Kashif and Saif Al-Kuwari. Resqnets: a residual approach for mitigating barren plateaus in quantum neural networks. EPJ Quantum Technology, 11(1):4, 2024. 4

2024
[20]

Resqunns: Towards enabling deep learning in quantum convolution neu- ral networks

Muhammad Kashif and Muhammad Shafique. Resqunns: Towards enabling deep learning in quantum convolution neu- ral networks. arXiv preprint arXiv:2402.09146, 2024. 1

2024 arXiv
[21]

Continuous-variable quantum neural networks

Nathan Killoran, Thomas R Bromley, Juan Miguel Ar- razola, Maria Schuld, Nicol ´as Quesada, and Seth Lloyd. Continuous-variable quantum neural networks. Physical Re- view Research, 1(3):033063, 2019. 1, 4

2019
[22]

Large- scale Quantum Reservoir Learning with an Analog Quantum Computer

Milan Kornja ˇca, Hong-Ye Hu, Chen Zhao, Jonathan Wurtz, Phillip Weinberg, Majd Hamdan, Andrii Zhdanov, Sergio H Cantu, Hengyun Zhou, Rodrigo Araiza Bravo, et al. Large- scale Quantum Reservoir Learning with an Analog Quantum Computer. arXiv preprint arXiv:2407.02553, 2024. 1, 3

2024 arXiv
[23]

Differentiable analog quantum computing for opti- mization and control

Jiaqi Leng, Yuxiang Peng, Yi-Ling Qiao, Ming Lin, and Xi- aodi Wu. Differentiable analog quantum computing for opti- mization and control. Advances in Neural Information Pro- cessing Systems, 35:4707–4721, 2022. 4, 6

2022
[24]

Quantum reinforcement learning during human decision-making

Ji-An Li, Daoyi Dong, Zhengde Wei, Ying Liu, Yu Pan, Franco Nori, and Xiaochu Zhang. Quantum reinforcement learning during human decision-making. Nature human be- haviour, 4(3):294–307, 2020. 1

2020
[25]

A hybrid quantum–classical neural network with deep residual learning

Yanying Liang, Wei Peng, Zhu-Jun Zheng, Olli Silv ´en, and Guoying Zhao. A hybrid quantum–classical neural network with deep residual learning. Neural Networks, 143:133–147,
[26]

Digital-analog Quan- tum Learning on Rydberg Atom Arrays

Jonathan Z Lu, Lucy Jiao, Kristina Wolinski, Milan Kornjaˇca, Hong-Ye Hu, Sergio Cantu, Fangli Liu, Su- sanne F Yelin, and Sheng-Tao Wang. Digital-analog Quan- tum Learning on Rydberg Atom Arrays. arXiv preprint arXiv:2401.02940, 2024. 3, 8

2024 arXiv
[27]

Digital–analog quantum learn- ing on rydberg atom arrays

Jonathan Z Lu, Lucy Jiao, Kristina Wolinski, Milan Ko- rnjaˇca, Hong-Ye Hu, Sergio Cantu, Fangli Liu, Susanne F Yelin, and Sheng-Tao Wang. Digital–analog quantum learn- ing on rydberg atom arrays. Quantum Science and Technol- ogy, 10(1):015038, 2024. 4

2024
[28]

Opportunities in quantum reservoir com- puting and extreme learning machines

Pere Mujal, Rodrigo Mart ´ınez-Pe˜na, Johannes Nokkala, Jorge Garc´ıa-Beni, Gian Luca Giorgi, Miguel C Soriano, and Roberta Zambrini. Opportunities in quantum reservoir com- puting and extreme learning machines. Advanced Quantum Technologies, 4(8):2100027, 2021. 1 10

2021
[29]

Neural schr ¨odinger equation: Physical law as deep neural network

Mitsumasa Nakajima, Kenji Tanaka, and Toshikazu Hashimoto. Neural schr ¨odinger equation: Physical law as deep neural network. IEEE Transactions on Neural Net- works and Learning Systems, 33(6):2686–2700, 2022. 4

2022
[30]

Disq: a novel quantum output state classification method on ibm quantum comput- ers using openpulse

Tirthak Patel and Devesh Tiwari. Disq: a novel quantum output state classification method on ibm quantum comput- ers using openpulse. InProceedings of the 39th International Conference on Computer-Aided Design, pages 1–9, 2020. 2

2020
[31]

OPTIC: A Practical Quantum Binary Classifier for Near-term Quan- tum Computers

Tirthak Patel, Daniel Silver, and Devesh Tiwari. OPTIC: A Practical Quantum Binary Classifier for Near-term Quan- tum Computers. In 2022 Design, Automation & Test in Europe Conference & Exhibition (DATE) , pages 334–339. IEEE, 2022. 2

2022
[32]

Quantum convolutional neural networks (qcnn) using deep learning for computer vision applications

Varadi Rajesh, Umesh Parameshwar Naik, et al. Quantum convolutional neural networks (qcnn) using deep learning for computer vision applications. In 2021 International confer- ence on recent trends on electronics, information, communi- cation & technology (RTEICT), pages 728–73...

2021
[33]

ProxiML: Building Machine Learning Classifiers for Photonic Quantum Computing

Aditya Ranjan, Tirthak Patel, Daniel Silver, Harshitta Gandhi, and Devesh Tiwari. ProxiML: Building Machine Learning Classifiers for Photonic Quantum Computing. In Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Op- e...

2024
[34]

MosaiQ: Quantum Generative Adversarial Networks for Image Generation on NISQ Computers

Daniel Silver, Tirthak Patel, William Cutler, Aditya Ranjan, Harshitta Gandhi, and Devesh Tiwari. MosaiQ: Quantum Generative Adversarial Networks for Image Generation on NISQ Computers. In Proceedings of the IEEE/CVF Inter- national Conference on Computer Vision, pages 7030–7039,
[35]

SliQ: Quantum Image Similarity Networks on Noisy Quantum Computers

Daniel Silver, Tirthak Patel, Aditya Ranjan, Harshitta Gandhi, William Cutler, and Devesh Tiwari. SliQ: Quantum Image Similarity Networks on Noisy Quantum Computers. In Proceedings of the AAAI Conference on Artificial Intelli- gence, pages 9846–9854, 2023. 1

2023
[36]

Quantumnat: quantum noise-aware training with noise injection, quantization and normalization

Hanrui Wang, Jiaqi Gu, Yongshan Ding, Zirui Li, Frederic T Chong, David Z Pan, and Song Han. Quantumnat: quantum noise-aware training with noise injection, quantization and normalization. In Proceedings of the 59th ACM/IEEE design automation conference, pages 1–6, 2022. 2

2022
[37]

Enhancing the expressivity of quantum neural networks with residual connections

Jingwei Wen, Zhiguo Huang, Dunbo Cai, and Ling Qian. Enhancing the expressivity of quantum neural networks with residual connections. Communications Physics , 7(1):220,
[38]

Non-native Quantum Generative Optimization with Adversarial Autoencoders

Blake A Wilson, Jonathan Wurtz, Vahagn Mkhitaryan, Michael Bezick, Sheng-Tao Wang, Sabre Kais, Vladimir M Shalaev, and Alexandra Boltasseva. Non-native Quantum Generative Optimization with Adversarial Autoencoders. arXiv preprint arXiv:2407.13830, 2024. 3

2024 arXiv
[39]

Aquila: QuEra’s 256-Qubit Neutral-Atom Quantum Com- puter

Jonathan Wurtz, Alexei Bylinskii, Boris Braverman, Jesse Amato-Grill, Sergio H Cantu, Florian Huber, Alexander Lukin, Fangli Liu, Phillip Weinberg, John Long, et al. Aquila: QuEra’s 256-Qubit Neutral-Atom Quantum Com- puter. arXiv preprint arXiv:2306.11727, 2023. 3, 7

2023 arXiv
[40]

Fashion- mnist: a novel image dataset for benchmarking machine learning algorithms

Han Xiao, Kashif Rasul, and Roland V ollgraf. Fashion- mnist: a novel image dataset for benchmarking machine learning algorithms. arXiv preprint arXiv:1708.07747, 2017. 7 11

2017 arXiv

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

https://www.kaggle.com/datasets/ uciml/pima-indians-diabetes-database, 2024

Pima Indians Diabetes Database. https://www.kaggle.com/datasets/ uciml/pima-indians-diabetes-database, 2024. Accessed: 2024-10-01. 7

work page 2024

[2] [2]

Quantum computing optimization technique for iot platform using modified deep residual approach

Rasha M Abd El-Aziz, Ahmed I Taloba, and Fahad A Al- ghamdi. Quantum computing optimization technique for iot platform using modified deep residual approach. Alexandria Engineering Journal, 61(12):12497–12509, 2022. 4

work page 2022

[3] [3]

AWS Aquila Interface

Amazon-Braket. AWS Aquila Interface. https : / / github . com / amazon - braket / amazon - braket - examples / blob / main / examples / analog _ hamiltonian _ simulation / 01 _ Introduction _ to _ Aquila . ipynb, 2024. Ac- cessed: 2024-11-01. 7

work page 2024

[4] [4]

Qure: Qubit re-allocation in noisy intermediate-scale quan- tum computers

Abdullah Ash-Saki, Mahabubul Alam, and Swaroop Ghosh. Qure: Qubit re-allocation in noisy intermediate-scale quan- tum computers. In Proceedings of the 56th Annual Design Automation Conference (DAC), pages 1–6, 2019. 2

work page 2019

[5] [5]

Measuring analytic gradients of general quantum evolution with the stochastic parameter shift rule

Leonardo Banchi and Gavin E Crooks. Measuring analytic gradients of general quantum evolution with the stochastic parameter shift rule. Quantum, 5:386, 2021. 4, 6

work page 2021

[6] [6]

Muqut: Multi-constraint quantum circuit mapping on nisq computers

Debjyoti Bhattacharjee, Abdullah Ash Saki, Mahabubul Alam, Anupam Chattopadhyay, and Swaroop Ghosh. Muqut: Multi-constraint quantum circuit mapping on nisq computers. In 2019 IEEE/ACM international conference on computer-aided design (ICCAD), pages 1–7. IEEE, 2019. 2

work page 2019

[7] [7]

Quantum ma- chine learning

Jacob Biamonte, Peter Wittek, Nicola Pancotti, Patrick Rebentrost, Nathan Wiebe, and Seth Lloyd. Quantum ma- chine learning. Nature, 549(7671):195–202, 2017. 1

work page 2017

[8] [8]

Challenges and opportu- nities in quantum machine learning

Marco Cerezo, Guillaume Verdon, Hsin-Yuan Huang, Lukasz Cincio, and Patrick J Coles. Challenges and opportu- nities in quantum machine learning. Nature Computational Science, 2(9):567–576, 2022. 1

work page 2022