Pith. sign in

REVIEW 4 major objections 4 minor 42 references

Q-SpiRL: Quantum Spiking Reinforcement Learning for Adaptive Robot Navigation

T0 review · 4 major / 4 minor · reviewed 2026-08-02 · deepseek-v4-flash

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

Pith's one-line read A hybrid quantum-spiking reinforcement learning agent, QSNN, is claimed to achieve the strongest balance of success rate, path efficiency, and motion smoothness in obstacle-aware grid navigation, with 99% success in a 40x40 environment.

desk verdict A cleanly written exploratory study that combines a variational quantum circuit with spiking RL for grid navigation, but its central claim that the quantum layer helps is confounded by differing hidden sizes and independently tuned hyperparameters. read the letter →

arxiv 2605.20801 v2 pith:AOAWEDAE submitted 2026-05-20 cs.RO quant-ph

Mohamed Khair Altrabulsi , Nouhaila Innan , Alberto Marchisio , Muhammad Kashif , Muhammad Shafique This is my paper · ORCID
classification cs.ROquant-ph
keywords quantumreinforcementlearningspikingneuralnetworksrobotnavigationvariationalcircuitsQ-tableconversiondynamicobstacleavoidancehybridquantum-classicalpolicygrid-world
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

This paper tries to establish that inserting a variational quantum circuit into a spiking neural-network policy improves reinforcement learning for obstacle-aware robot navigation. It presents Q-SpiRL, a framework that trains five agent families — tabular Q-learning, classical MLP, classical spiking (SNN), quantum-enhanced MLP (QMLP), and quantum-enhanced spiking (QSNN) — under a shared environment and evaluation protocol. The central claim is that QSNN achieves the best overall trade-off among task completion, path efficiency, and motion smoothness across 20x20, 30x30, and 40x40 grids, reaching 99% success in the hardest setting while keeping high success-weighted path length and low turn rate. A sympathetic reader would care because the paper offers a concrete, tested mechanism for hybrid quantum policies — spike-derived firing rates feeding a quantum feature transformation — and shows a single successful deployment on real quantum hardware.

What carries the argument

The central mechanism is a hybrid QSNN pipeline: a frequency-based Poisson encoder converts the one-hot state into spike trains; leaky integrate-and-fire (LIF) layers process the trains over time; mean-pooling produces an 8-dimensional firing-rate vector; a variational quantum circuit (Hadamard gates, trainable controlled rotations, entangling layers, and Pauli-Z measurements) transforms that vector into quantum features; a final classical layer maps them to five Q-values. The secondary mechanism is the conversion of every trained policy into an explicit Q-table over the 2,560 discrete states, so all five agents are compared using the same deterministic greedy lookup.

What would settle it

Retrain all five agents with the state extended to include the agent's absolute position and distance to the target, keeping the rest of the protocol identical. If QSNN no longer outperforms the classical SNN, or if a brute-force check finds two states with equal four-component features but different optimal actions, the central claim about quantum-enhanced spiking navigation is undermined.

Watch

Extended reading notes

Core claim

The central discovery claimed by the paper is that the QSNN — a spiking Q-network whose temporally averaged firing rates are passed through an eight-qubit, three-layer variational quantum circuit before action-value estimation — gives the strongest joint performance among the five evaluated families. In all three grid sizes, QSNN matches or exceeds the classical SNN on every metric; at 40x40 it reaches 99% success with SPL 0.902 and turn rate 0.260, while the classical MLP falls to 77% success. The paper interprets this as evidence that spike-based temporal representations are an effective pre-quantum interface, and that quantum enhancement is especially beneficial in the spiking pipeline ra

Load-bearing premise

The load-bearing assumption is that the four-component state (Ro, Do, RT, AT→o) fully captures what the agent needs to know, so a 2,560-entry lookup table can represent an optimal policy; if position and target-distance are necessary, the Q-table conversion cannot express the true optimal policy and the headline numbers rest on a partial state.

Editorial extensions

If this is right

  • Quantum enhancement helps the spiking agent uniformly across all four metrics, while dense MLPs gain less and degrade sharply at 40x40.
  • Spiking policies remain more robust than dense policies as environment size and obstacle complexity grow.
  • The explicit Q-table conversion makes deployment inference a lookup operation and lets stochastic or neural policies be evaluated deterministically and comparably.
  • The successful single episode on real quantum hardware is a feasibility proof, though the paper explicitly says it is not a statistically matched benchmark.
  • Success rate alone can mislead: the tabular Q-learning baseline often matches success rate but trails on SPL and turn rate, so quality metrics are needed.

Reading between the lines

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

  • Because the state vector leaves out the agent's position and its distance to the goal, the policy cannot distinguish scenes that look identical in relative angles but differ in absolute location. A natural test is to add position and target distance to the state and see whether QSNN still beats the classical SNN; if the margin disappears, the reported quantum advantage depends on the specific disc
  • The Q-table construction for spiking agents uses one sampled Poisson-encoding realization per state. Repeating the conversion with different seeds would show how sensitive the reported QSNN numbers are to that single draw.
  • The hardware section labels its experiment as an indicative feasibility check and reports only one episode; averaging many hardware episodes would convert that check into an estimate of the noise-induced drop in SPL and reveal whether near-term devices can sustain the advantage.
  • The angular-bin state representation suggests a transfer path to real robots: quantize continuous range measurements into the same bins; whether the quantum layer still helps after that quantization would test the mechanism beyond grid worlds.
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

4 major / 4 minor

Summary. The paper presents Q-SpiRL, a framework that combines spiking reinforcement learning with a variational quantum circuit for grid-world robot navigation. It evaluates five agent families (tabular Q-learning, MLP, SNN, QMLP, QSNN) in 20x20, 30x30, and 40x40 environments with static and dynamic obstacles. The central claim is that the QSNN, which processes spike-derived firing rates through a parameterized quantum circuit, achieves the strongest overall trade-off among success rate, SPL, path length, and turn rate, and that the quantum layer consistently improves over the classical SNN. The paper also reports a single-episode feasibility run on IBM quantum hardware.

Significance. If the central comparison were valid, the paper would make a useful empirical contribution to hybrid quantum-spiking reinforcement learning, with a clean deterministic evaluation protocol and a hardware feasibility demonstration. The unified Q-table conversion and the use of multiple navigation metrics are commendable. However, the key causal claim—that the variational quantum layer improves spiking RL policies—is not supported by the current experiments because the QSNN and SNN differ in hidden-neuron count, hyperparameters, and parameter count. In addition, the compact state representation in Eq. (7) is not Markovian, which undermines the validity of the Q-table evaluation as a representation of an optimal policy. These are load-bearing issues that prevent acceptance.

major comments (4)
  1. [Section III.B.4, Table I, Table II] The QSNN-vs-SNN comparison does not isolate the quantum layer. Table II fixes classical hidden neurons at 30 and quantum hidden neurons at 35, and Table I selects different spiking hyperparameters for the classical and quantum settings in every environment (e.g., 40×40: fmax 200 vs 100 Hz, T 5 vs 20, Δt 0.01 vs 0.20). Parameter counts also differ (QSNN 1,977 vs SNN 1,985), contradicting the 'parameter-matched' description. The Table III gains could therefore be due to network capacity or temporal dynamics rather than the variational circuit. A matched ablation that changes only the quantum layer is required for the central claim.
  2. [Section III.A, Eq. (7)] The state st=(Ro,Do,RT,AT→o) is not a sufficient statistic for the navigation task: it omits the agent position, the distance to the target, the distance to the nearest obstacle, and the dynamic obstacle's position. The reward in Eq. (9) explicitly depends on Euclidean distances, so the process is partially observable. Consequently, the Q-table conversion in Section III.D over the 2,560 states cannot represent an optimal Markovian policy, and all Table III metrics are for a reactive POMDP policy. The paper must either augment the state with the missing coordinates/distances or provide a formal justification of Markovianity.
  3. [Section IV.B, Table III] No statistical significance testing is reported, and many of the claimed QSNN-vs-SNN differences are within the reported standard errors. For example, in 20×20 the SPL values are 0.8933±0.0192 vs 0.8765±0.0211; in 40×40 the success rates are 0.990±0.0100 vs 0.980±0.0141. The repeated statement that the quantum layer provides 'consistent gains' is not supported without paired tests over the same 100 seeds or confidence intervals for the differences.
  4. [Section III.C] The hyperparameter selection protocol is a selection-on-performance procedure: each family is tuned independently, configurations with success rate below 95% are discarded, and the best path length among the survivors is chosen before reporting the final metrics. This creates an optimism bias and, combined with the small evaluation set, can inflate apparent advantages. The paper should separate training, validation, and test splits, and report the selection outcomes for all searched configurations.
minor comments (4)
  1. [Section III.A, Eq. (9)] The reward coefficients β1, β2, β3 and the magnitude of the per-step penalty are never specified. Without these values the experiments cannot be reproduced.
  2. [Section III.A] The phrase 'one additional training-only diagonal static obstacle' is ambiguous: if this obstacle is absent at test time, the training and test distributions differ; if it is present at test time, 'training-only' is misleading. Please clarify.
  3. [Section III.F, Eq. (23)] The grid-cell side length Δ is not defined, so the reported path lengths in meters are not reproducible. Please state the physical scale used.
  4. [Table II] The table lists an initial temperature T0 and temperature decay for exploration, while the text states ε-greedy with ε=0.01. Please clarify which exploration scheme is actually used.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation; the QSNN advantage is an empirical comparison, not an identity or a self-citation chain.

full rationale

The paper's central claim is that QSNN provides the strongest overall trade-off among the evaluated agents. This claim is an observed result from Table III under a fixed evaluation protocol, not a quantity derived from its own inputs. The only potentially circular-looking elements are (i) the independent per-family hyperparameter search and (ii) self-citations in the related-work section. Neither is circular by the standards required here. Hyperparameters are selected using training success rate and mean successful path length, while reported test metrics come from held-out seeds Ztest={2000,...,2099}; the comparison is not forced by the selection objective. SPL and turn rate are not part of the selection criterion, and test environments are distinct from the grid-search data. The different hidden widths (30 vs. 35) and spiking hyperparameters in Tables I and II are a confound for the causal statement that the quantum layer improves SNN performance, but a confounded comparison is not an equivalence-by-construction. Self-citations [40] and [42] appear only as background for quantum spiking networks and are not invoked as proof of QSNN effectiveness. The hardware result in Section IV.D is explicitly labeled as a single-episode feasibility check. No equation defines QSNN's reported performance in terms of the input state or fitted hyperparameters, and no fitted parameter is renamed as a prediction. The derivation chain, insofar as one exists, is an empirical pipeline rather than a closed loop.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The paper introduces no new physical entities or forces. Its free parameters are reward-shaping constants, obstacle-generation counts, and per-environment hyperparameters; the load-bearing ad hoc assumption is that the 4-component state vector is Markovian.

free parameters (4)
  • Reward shaping coefficients β1, β2, β3 = not reported
    Eq. (9) defines reward as β1(dt-1-dt) + exp(β2(d_ot-1-d_ot)) + β3 sin(αt). The paper never gives their values; they are hand-selected weights that directly shape the policy and, if changed, would alter all results.
  • Per-step penalty = not reported
    Section III.A mentions 'a small per-step penalty is applied' without specifying its magnitude; this parameter trades path length against success and is central to the path-efficiency claims.
  • Obstacle counts and 'training-only diagonal static obstacle' = 6/9/12 static; 1/1/3 dynamic; +1 training-only diagonal
    Section III.A chooses these numbers by hand. The training-only diagonal obstacle is an ad hoc addition to force non-trivial behavior, and the random distribution of obstacle placement is not specified, making the benchmark hard to reproduce.
  • Spiking/quantum hyperparameters per environment = see Table I (e.g., 30x30: SNN T=5, Δt=0.05; QSNN T=20, Δt=0.05; 40x40: SNN fmax=100, QSNN fmax=200)
    Section III.C grid-searches fmax, T, Δt, τmem, τsyn independently for each model family and each environment. These fitted choices mean the QSNN vs SNN comparison is not controlled for hyperparameters.
assumptions (5)
  • ad hoc to paper Eq. (7) state vector is a sufficient statistic for optimal decision-making
    The policy depends only on Ro, Do, RT, AT→o; position and distance to target are absent. If the MDP is not Markovian under this state, the Q-table over the 2,560 states cannot represent the optimal policy.
  • domain assumption LIF neuron model and Poisson frequency encoding faithfully represent the intended spiking computation
    Section III.B.2 uses standard leaky integrate-and-fire neurons and a Poisson encoder; the paper does not validate that the discretization steps (Eq. 10, 11) preserve the intended temporal dynamics.
  • standard math Tabular Q-learning converges to the optimal Q-function in the finite MDP
    Invoked implicitly in Section II.A and used to justify the tabular baseline; assumes standard convergence conditions (sufficient exploration, learning-rate schedule) hold.
  • domain assumption A single Poisson encoding realization per state is representative when constructing the Q-table
    Section III.D states 'A single offline encoding realization is used'; this stochasticity is confined to table construction but its effect on the resulting greedy policy is not quantified.
  • domain assumption Fixed environment constants (Tmax=300, dynamic obstacle update every 2 steps, action set with Δθ ∈ {±π/2, ±π/4, 0}) define a meaningful navigation task
    These are chosen by hand in Section III.A and are not varied; the agent's performance is only meaningful under these specific choices.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Q-SpiRL: Quantum Spiking Reinforcement Learning for Adaptive Robot Navigation." pith.science (2026). https://pith.science/paper/AOAWEDAE

@misc{pith2026260520801,
  author       = {Pith},
  title        = {Pith review of: Q-SpiRL: Quantum Spiking Reinforcement Learning for Adaptive Robot Navigation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AOAWEDAE}},
  note         = {Machine review of arXiv:2605.20801}
}
read the original abstract

Adaptive robot navigation in dynamic environments requires policies that can reach the target reliably while producing efficient and stable trajectories. This paper presents Q-SpiRL, a quantum spiking reinforcement learning framework for obstacle-aware robot navigation. The framework develops and evaluates five agent families: tabular Q-learning, classical MLP, classical SNN, quantum-enhanced MLP (QMLP), and quantum-enhanced spiking neural network (QSNN). While all models are implemented under a unified training and evaluation pipeline, the QSNN is the central architecture of interest, as it combines spike-based temporal processing with variational quantum feature transformation. Experiments are conducted across three grid-world environments of increasing size, namely 20x20, 30x30, and 40x40, with both static and dynamic obstacles. Performance is assessed using success rate, success-weighted path length, path length, and turn rate under deterministic inference. Results show that QSNN achieves the strongest overall trade-off between task completion, trajectory efficiency, and motion smoothness, reaching up to 99% success rate while maintaining high path efficiency in the most challenging setting. Execution on IBM quantum hardware further demonstrates the feasibility of deploying the proposed hybrid policy under real-device conditions.

Figures

Figures reproduced from arXiv: 2605.20801 by the authors.

Figure 1
Figure 1. Overview of the proposed methodology pipeline. The framework begins with the definition of the navigation environment. It [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Architecture of the classical MLP. 2) Classical SNN: The classical SNN replaces conventional pointwise nonlinearities with leaky integrate-and-fire (LIF) neurons and processes the encoded state through temporal spike dynamics. As in the MLP case, the discrete state is first transformed into the same 29-dimensional one-hot representation and is then converted into a spike-based input sequence using a frequency-based … view at source ↗
Figure 3
Figure 3. Architecture of the classical SNN. 3) Quantum-Enhanced MLP (QMLP): The quantum￾enhanced multilayer perceptron extends the classical MLP by inserting a variational quantum circuit between classical preprocessing and output layers. The one-hot encoded input vector x ∈ R 29 is first projected by classical fully connected layers into a latent feature vector whose dimension matches the number of qubits, with q = 8. This … view at source ↗
Figures from the paper (2 more)
Figure 5
Figure 5. Figure 5: illustrates representative QSNN trajectories in the 20×20, 30×30, and 40×40 environments. The figure shows that the learned policy maintains coherent obstacle-aware navigation across all tested scales, supporting the quantitative evidence that the QSNN preserves both e…
Figure 6
Figure 6. Figure 6: shows the trajectory obtained on hardware. Despite non-ideal execution conditions, the policy generates a valid obstacle-avoiding path and reaches the goal. These results provide an initial proof of feasibility for deploying the proposed quantum-enhanced spiking policy…

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references · 5 linked inside Pith

  1. [1]

    Vision for mobile robot navigation: A survey,

    G. N. DeSouza and A. C. Kak, “Vision for mobile robot navigation: A survey,”IEEE transactions on pattern analysis and machine intelligence, vol. 24, no. 2, pp. 237–267, 2002

    2002

  2. [2]

    A survey of autonomous robots and multi-robot naviga- tion: Perception, planning and collaboration,

    W. Chenet al., “A survey of autonomous robots and multi-robot naviga- tion: Perception, planning and collaboration,”Biomimetic Intelligence and Robotics, vol. 5, no. 2, p. 100203, 2025

    2025

  3. [3]

    A comprehensive review on autonomous navigation,

    S. Nahavandiet al., “A comprehensive review on autonomous navigation,” ACM Computing Surveys, vol. 57, no. 9, pp. 1–67, 2025

    2025

  4. [4]

    R. S. Sutton, A. G. Bartoet al.,Reinforcement learning: An introduction. MIT press Cambridge, 1998, vol. 1, no. 1

    1998

  5. [5]

    Deep reinforcement learning based mobile robot navigation: A review,

    K. Zhu and T. Zhang, “Deep reinforcement learning based mobile robot navigation: A review,”Tsinghua Science and Technology, vol. 26, no. 5, pp. 674–691, 2021

    2021

  6. [6]

    Q-learning,

    C. J. Watkins and P. Dayan, “Q-learning,”Machine learning, vol. 8, no. 3, pp. 279–292, 1992

    1992

  7. [7]

    A review of reinforcement learning for autonomous building energy management,

    K. Mason and S. Grijalva, “A review of reinforcement learning for autonomous building energy management,”Computers & Electrical Engineering, vol. 78, pp. 300–312, 2019

    2019

  8. [8]

    Reinforcement learning, fast and slow,

    M. Botvinick, S. Ritter, J. X. Wang, Z. Kurth-Nelson, C. Blundell, and D. Hassabis, “Reinforcement learning, fast and slow,”Trends in cognitive sciences, vol. 23, no. 5, pp. 408–422, 2019

    2019

Show all 42 references
  1. [9]

    Spiking neural networks,

    S. Ghosh-Dastidar and H. Adeli, “Spiking neural networks,”International journal of neural systems, vol. 19, no. 04, pp. 295–308, 2009

    2009

  2. [10]

    A survey of robotics control based on learning-inspired spiking neural networks,

    Z. Bing, C. Meschede, F. R ¨ohrbein, K. Huang, and A. C. Knoll, “A survey of robotics control based on learning-inspired spiking neural networks,”Frontiers in neurorobotics, vol. 12, p. 35, 2018

    2018

  3. [11]

    A systematic review of spiking neural networks for human-robot interaction in rehabilitative wearable robotics,

    X. Zhang, Y . Cao, J. Huang, J. Liu, and Z.-Q. Zhang, “A systematic review of spiking neural networks for human-robot interaction in rehabilitative wearable robotics,”IEEE Transactions on Cognitive and Developmental Systems, 2025

    2025

  4. [12]

    Deep reinforcement learning with spiking q-learning,

    D. Chen, P. Peng, T. Huang, and Y . Tian, “Deep reinforcement learning with spiking q-learning,”ArXiv, vol. abs/2201.09754, 2022

    2022 arXiv

  5. [13]

    Exploring spiking neural networks for deep reinforcement learning in robotic tasks,

    L. Zanatta, F. Barchi, S. Manoni, S. Tolu, A. Bartolini, and A. Acquaviva, “Exploring spiking neural networks for deep reinforcement learning in robotic tasks,”Scientific Reports 2024 14:1, vol. 14, pp. 30 648–, 12 2024

    2024

  6. [14]

    Dsqn: Robust path planning of mobile robot based on deep spiking q-network,

    A. Kumar, L. Zhang, H. Bilal, S. Wang, A. M. Shaikh, L. Bo, A. Rohra, and A. Khalid, “Dsqn: Robust path planning of mobile robot based on deep spiking q-network,”Neurocomputing, vol. 634, p. 129916, 2025

    2025

  7. [15]

    Quantum machine learning,

    J. Biamonte, P. Wittek, N. Pancotti, P. Rebentrost, N. Wiebe, and S. Lloyd, “Quantum machine learning,”Nature 2017 549:7671, vol. 549, pp. 195– 202, 9 2017

    2017

  8. [16]

    A primer on quantum machine learning,

    S. Y . Chang and M. Cerezo, “A primer on quantum machine learning,” arXiv preprint arXiv:2511.15969, 2025

    2025

  9. [17]

    The emergence of deep reinforcement learning for path planning,

    T. T. Nguyen, S. Nahavandi, I. Razzak, D. Nguyen, N. T. Pham, and Q. Viet Hung Nguyen, “The emergence of deep reinforcement learning for path planning,” in2025 IEEE International Conference on Systems, Man, and Cybernetics (SMC), 2025, pp. 6265–6272

    2025

  10. [18]

    Exploring neuromorphic computing based on spiking neural networks: Algorithms to hardware,

    N. Rathi, I. Chakraborty, A. Kosta, A. Sengupta, A. Ankit, P. Panda, and K. Roy, “Exploring neuromorphic computing based on spiking neural networks: Algorithms to hardware,”ACM Computing Surveys, vol. 55, no. 12, pp. 1–49, 2023

    2023

  11. [19]

    Hsrl: A hierarchical control system based on spiking deep reinforcement learning for robot navigation,

    B. Yanget al., “Hsrl: A hierarchical control system based on spiking deep reinforcement learning for robot navigation,”Proceedings - IEEE International Conference on Robotics and Automation, 2025

    2025

  12. [20]

    Lep-qnn: Loan eligibility prediction using quantum neural networks,

    N. Innan, A. Marchisio, M. Bennai, and M. Shafique, “Lep-qnn: Loan eligibility prediction using quantum neural networks,” in2025 IEEE International Conference on Quantum Computing and Engineering (QCE), vol. 1. IEEE, 2025, pp. 1864–1872

    2025

  13. [21]

    Qnn-vrcs: A quantum neural network for vehicle road cooperation systems,

    N. Innan, B. K. Behera, S. Al-Kuwari, and A. Farouk, “Qnn-vrcs: A quantum neural network for vehicle road cooperation systems,”IEEE Transactions on Intelligent Transportation Systems, 2025

    2025

  14. [22]

    Next- generation quantum neural networks: Enhancing efficiency, security, and privacy,

    N. Innan, M. Kashif, A. Marchisio, M. Bennai, and M. Shafique, “Next- generation quantum neural networks: Enhancing efficiency, security, and privacy,” in2025 IEEE 31st International Symposium on On-Line Testing and Robust System Design (IOLTS). IEEE, 2025, pp. 1–4

    2025

  15. [23]

    Parametrized quantum policies for reinforcement learning,

    S. Jerbi, C. Gyurik, S. Marshall, H. Briegel, and V . Dunjko, “Parametrized quantum policies for reinforcement learning,”Advances in neural information processing systems, vol. 34, pp. 28 362–28 375, 2021

    2021

  16. [24]

    Qadqn: Quantum attention deep q-network for financial market prediction,

    S. Dutta, N. Innan, A. Marchisio, S. B. Yahia, and M. Shafique, “Qadqn: Quantum attention deep q-network for financial market prediction,” in2024 IEEE International Conference on Quantum Computing and Engineering (QCE), vol. 2. IEEE, 2024, pp. 341–346

    2024

  17. [25]

    QAS-QTNs: Curriculum reinforcement learning-driven quantum architecture search for quantum tensor networks,

    S. Dutta, N. Innan, S. B. Yahia, and M. Shafique, “QAS-QTNs: Curriculum reinforcement learning-driven quantum architecture search for quantum tensor networks,” in2025 IEEE International Conference on Quantum Computing and Engineering (QCE), vol. 1. IEEE, 2025, pp. 1739–1747

    2025

  18. [26]

    A survey on quantum reinforcement learning,

    N. Meyer, C. Ufrecht, M. Periyasamy, D. D. Scherer, A. Plinge, and C. Mutschler, “A survey on quantum reinforcement learning,”arXiv preprint arXiv:2211.03464, 2022

    2022 arXiv

  19. [27]

    Quantum reinforcement learning in dynamic environments,

    O. Sefrin, M. Radons, L. Simon, and S. W ¨olk, “Quantum reinforcement learning in dynamic environments,”arXiv preprint arXiv:2507.01691, 2025

    2025 arXiv

  20. [28]

    Robust quantum-inspired reinforcement learning for robot navigation,

    D. Dong, C. Chen, J. Chu, and T.-J. Tarn, “Robust quantum-inspired reinforcement learning for robot navigation,”IEEE/ASME transactions on mechatronics, vol. 17, no. 1, pp. 86–97, 2010

    2010

  21. [29]

    Quantum deep reinforcement learning for robot navigation tasks,

    H. Hohenfeld, D. Heimann, F. Wiebe, and F. Kirchner, “Quantum deep reinforcement learning for robot navigation tasks,”IEEE Access, vol. 12, pp. 87 217–87 236, 2024

    2024

  22. [30]

    Nav-q: quantum deep reinforce- ment learning for collision-free navigation of self-driving cars,

    A. Sinha, A. Macaluso, and M. Klusch, “Nav-q: quantum deep reinforce- ment learning for collision-free navigation of self-driving cars,”Quantum Machine Intelligence, vol. 7, no. 1, p. 19, 2025

    2025

  23. [31]

    Quantum-enhanced hybrid reinforcement learning framework for dynamic path planning in autonomous systems,

    S. Tomar, S. Alam, S. Kumar, and A. Mathur, “Quantum-enhanced hybrid reinforcement learning framework for dynamic path planning in autonomous systems,”arXiv preprint arXiv:2504.20660, 2025

    2025 arXiv

  24. [32]

    Qmarl: A quantum multi- agent reinforcement learning framework for swarm robots navigation,

    W. Chen, J. Wan, F. Ye, R. Wang, and C. Xu, “Qmarl: A quantum multi- agent reinforcement learning framework for swarm robots navigation,” in2024 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW). IEEE, 2024, pp. 388–392

    2024

  25. [33]

    QUA V: Quantum-assisted path planning and optimization for uav navigation with obstacle avoidance,

    N. Innan, M. Kashif, A. Marchisio, Y .-S. Gan, F. Barbaresco, and M. Shafique, “QUA V: Quantum-assisted path planning and optimization for uav navigation with obstacle avoidance,” in2025 IEEE International Conference on Quantum Artificial Intelligence (QAI). IEEE, 2025, pp. 208–215

    2025

  26. [34]

    A novel hybrid quantum architecture for path planning in quantum-enabled autonomous mobile robots,

    M. Sarkar, J. Pradhan, A. K. Singh, and H. Nenavath, “A novel hybrid quantum architecture for path planning in quantum-enabled autonomous mobile robots,”IEEE Transactions on Consumer Electronics, vol. 70, no. 3, pp. 5597–5606, 2024

    2024

  27. [35]

    Quantum robotics: a review of emerging trends,

    F. Yan, A. M. Iliyasu, N. Li, A. S. Salama, and K. Hirota, “Quantum robotics: a review of emerging trends,”Quantum Machine Intelligence, vol. 6, no. 2, p. 86, 2024

    2024

  28. [36]

    Current trends and advances in quantum navigation for maritime applications: A comprehensive review,

    O. Sambataro, R. Costanzi, J. Alves, A. Caiti, P. Paglierani, R. Petroccia, and A. Munaf `o, “Current trends and advances in quantum navigation for maritime applications: A comprehensive review,”IEEE Journal of Oceanic Engineering, 2025

    2025

  29. [37]

    Classification with integrated quantum and spiking neural networks,

    D. Pasquali, M. Grossi, and S. Vallecorsa, “Classification with integrated quantum and spiking neural networks,” in2023 IEEE International Conference on Quantum Computing and Engineering (QCE), vol. 2. IEEE, 2023, pp. 298–299

    2023

  30. [38]

    A quantum leaky integrate-and-fire spiking neuron and network,

    D. Brand and F. Petruccione, “A quantum leaky integrate-and-fire spiking neuron and network,”npj Quantum Information 2024 10:1, vol. 10, pp. 125–, 12 2024

    2024

  31. [39]

    Quantum-enhanced spiking neural networks,

    R. Khatoniar, D. Konar, and V . Aggarwal, “Quantum-enhanced spiking neural networks,”Proceedings - IEEE Quantum Week 2024, QCE 2024, vol. 2, pp. 490–491, 2024

    2024

  32. [40]

    FL-QDSNNs: Federated learning with quantum dynamic spiking neural networks,

    N. Innan, A. Marchisio, and M. Shafique, “FL-QDSNNs: Federated learning with quantum dynamic spiking neural networks,” in2025 IEEE International Conference on Quantum Artificial Intelligence (QAI). IEEE, 2025, pp. 113–119

    2025

  33. [41]

    Quantum spiking neural networks for image classification,

    S. Liu and Y . Gu, “Quantum spiking neural networks for image classification,” inThird International Conference on Algorithms, Network, and Communication Technology (ICANCT 2024), vol. 13545. SPIE, 2025, pp. 181–188

    2024

  34. [42]

    Spate: Spiking-phase adaptive temporal encoding for quantum machine learning,

    N. Innan, R. V . W. Putra, and M. Shafique, “Spate: Spiking-phase adaptive temporal encoding for quantum machine learning,”arXiv preprint arXiv:2604.11022, 2026

    2026 arXiv

Pith tools

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