pith. sign in

arxiv: 2607.01080 · v1 · pith:NXID63IVnew · submitted 2026-07-01 · 💻 cs.LG · cs.IT· math.IT

Balancing Expressivity and Learnability in Quantum Kernel Bandit Optimization

Yuqi Huang , Vincent Y. F. Tan , Sharu Theresa Jose This is my paper

Pith reviewed 2026-07-02 15:51 UTC · model grok-4.3

classification 💻 cs.LG cs.ITmath.IT
keywords quantum kernelsgaussian process banditsregret boundsprojected kernelsmisspecified GPkernel approximationquantum optimizationNISQ applications
0
0 comments X

The pith

Projected quantum kernels and classical approximations reduce dimensionality in GP bandit optimization while preserving key quantum properties, yielding misspecified algorithms with regret bounds that guide model complexity.

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

This paper examines Gaussian process bandit optimization with quantum kernels for NISQ-era tasks such as quantum control and variational algorithms, assuming the mean reward lies in the quantum kernel's RKHS. Full high-dimensional kernels increase model complexity and information gain, which raises cumulative regret and harms learnability. The authors introduce projected quantum kernels and classical kernel approximations that lower feature dimensionality while retaining essential quantum traits. These approximations support misspecified GP bandit algorithms whose derived regret bounds quantify the trade-off between approximation error and information gain. The bounds supply concrete guidance on choosing model complexity, and experiments show the methods achieve better sample efficiency and lower overhead than full quantum kernels.

Core claim

Projected quantum kernels and classical kernel approximation techniques reduce feature dimensionality while preserving key quantum properties; using these approximate kernels yields misspecified GP bandit algorithms whose regret bounds characterize the trade-off between approximation error and information gain, thereby guiding selection of optimal model complexity and empirically outperforming full quantum kernels in sample efficiency while cutting computational overhead.

What carries the argument

Projected quantum kernels and classical kernel approximation techniques that reduce feature dimensionality while preserving key quantum properties.

If this is right

  • Regret bounds quantify the trade-off between approximation error and information gain.
  • The bounds give explicit guidance for selecting optimal model complexity.
  • Misspecified GP algorithms achieve improved sample efficiency over full quantum kernels.
  • Computational overhead is substantially reduced, enabling scalable optimization.
  • The approach supports GP optimization for quantum-native applications on NISQ hardware.

Where Pith is reading between the lines

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

  • The same projection and approximation strategy could be tested on other kernel-based quantum algorithms beyond bandits.
  • The regret analysis framework may extend to settings where the mean function only approximately lies in the RKHS.
  • Empirical validation on actual quantum hardware would test whether the preserved quantum properties translate to practical advantage.
  • The dimensionality-reduction idea might inform kernel design choices in classical high-dimensional bandit problems.

Load-bearing premise

The mean reward function lies in the RKHS induced by the quantum kernel, and the proposed projections and approximations preserve the key quantum properties without introducing uncontrolled error that would invalidate the regret analysis.

What would settle it

A controlled experiment on a quantum control task in which the cumulative regret achieved by the projected or approximated kernels exceeds the regret of the full quantum kernel, or in which observed performance deviates from the predicted error-information-gain trade-off in the regret bounds.

Figures

Figures reproduced from arXiv: 2607.01080 by Sharu Theresa Jose, Vincent Y. F. Tan, Yuqi Huang.

Figure 1
Figure 1. Figure 1: Cumulative regrets over T = 100 rounds (us￾ing the EC-GP-UCB algorithm) with three different mod￾elling kernels on a synthetic task where the unknown re￾ward function lies in a quantum RKHS. While the full quan￾tum kernel beats the classical RBF kernel, our proposed projected quantum kernel-based approximation achieves an even smaller regret. See subsection 4.1 for experimen￾tal details. modelling the mean… view at source ↗
Figure 2
Figure 2. Figure 2: Plot of γT as a function of (a) the iterations T for varying system qubit size n, with Pauli-Y and Z rotation encoding feature maps, and (b) the number n of qubits with T = 1000, verifying the exponential increase of γT with n. approximation approaches: (a) projected quantum kernels via reduced quantum circuits; (b) classical Random Fourier Features (RFFs) for efficient kernel approximation; and (c) P-gree… view at source ↗
Figure 3
Figure 3. Figure 3: Cumulative regret of SquareCB against approximation model complexity for [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Cumulative regret of EC-GP-UCB against approximation model complexity for [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Uniform (max-norm) approximation error against [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Cumulative regrets of EC-GP-UCB after T = 100 iterations as a function of the model complexity for (a) projected quantum kernels and (b) P-greedy approximation for quantum phase classification. 1    [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Bayesian optimization of VQE with different ker [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Example Circuit Structure for generating [PITH_FULL_IMAGE:figures/full_fig_p037_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Cumulative regret of SquareCB algorithm as a function of approximation dimension [PITH_FULL_IMAGE:figures/full_fig_p039_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: EC-GP-UCB with different kernel approximation approaches for reward functions drawn from GPs with the [PITH_FULL_IMAGE:figures/full_fig_p039_10.png] view at source ↗
read the original abstract

We investigate Gaussian process (GP) bandit optimization with quantum kernels, assuming the mean reward function lies in the reproducing kernel Hilbert space (RKHS) induced by the quantum kernel. This setting is motivated by NISQ-era tasks such as quantum control, state preparation and variational quantum algorithms. While quantum kernels can offer a `quantum advantage' via domain-specific inductive biases, na\"{i}vely using full, high-dimensional kernels increases model complexity and information gain, leading to higher cumulative regret and poor learnability. To address this, we propose projected quantum kernels and classical kernel approximation techniques that reduce feature dimensionality while preserving key quantum properties. Using these approximate kernels, we develop misspecified GP bandit algorithms and derive regret bounds that characterize the trade-off between approximation error and information gain. The regret bounds provide principled guidance for selecting the optimal model complexity. Empirically, our methods outperform full quantum kernels in sample efficiency, while substantially reducing computational overhead, enabling scalable GP optimization for quantum-native applications.

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

0 major / 3 minor

Summary. The paper investigates GP bandit optimization with quantum kernels under the assumption that the mean reward lies in the induced RKHS. It proposes projected quantum kernels and classical approximations to reduce feature dimensionality while preserving key quantum properties, develops misspecified GP bandit algorithms, and derives regret bounds characterizing the expressivity-learnability trade-off to guide model complexity selection. Empirical results claim improved sample efficiency and reduced overhead versus full quantum kernels for NISQ tasks such as quantum control.

Significance. If the regret bounds hold with controlled approximation error, the work provides a principled framework for deploying quantum kernels in bandit settings without excessive model complexity, which is relevant for NISQ-era applications. The explicit trade-off characterization via regret analysis and the reported empirical gains in sample efficiency represent a concrete step toward scalable quantum kernel methods.

minor comments (3)
  1. The abstract states that regret bounds are derived but the provided text supplies no equation numbers or proof outlines; if the full manuscript contains them in §3–4, cross-reference them explicitly in the abstract for clarity.
  2. Notation for the projected kernel (likely defined in §2) should include an explicit equation showing how the projection operator acts on the feature map to make the dimensionality reduction precise.
  3. In the empirical section, the overhead reduction is claimed but no table or figure quantifies wall-clock time or memory versus the full kernel; adding such a comparison would strengthen the practical claim.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their positive assessment of our work on projected quantum kernels and misspecified GP bandit algorithms, and for recommending minor revision. The provided referee summary accurately reflects the manuscript's focus on the expressivity-learnability trade-off in quantum kernel bandit optimization for NISQ tasks.

Circularity Check

0 steps flagged

No significant circularity

full rationale

The derivation begins from the standard assumption that the mean reward lies in the RKHS of the quantum kernel, then introduces projected kernels and approximations whose error is controlled in the regret analysis. Regret bounds are derived to quantify the resulting expressivity-learnability trade-off, and empirical comparisons are presented separately. No load-bearing step reduces by construction to a fitted parameter, self-citation, or renamed input; the central claims rest on external GP bandit theory and explicit approximation-error terms rather than self-referential definitions.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract supplies no explicit free parameters, axioms, or invented entities; all such elements are therefore recorded as empty.

pith-pipeline@v0.9.1-grok · 5706 in / 934 out tokens · 23230 ms · 2026-07-02T15:51:10.444504+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

76 extracted references · 22 canonical work pages · 5 internal anchors

  1. [1]

    arXiv preprint arXiv:1506.02785 , year=

    On the error of random Fourier features , author=. arXiv preprint arXiv:1506.02785 , year=

    work page arXiv

  2. [2]

    arXiv preprint arXiv:2502.04021 , year=

    Variational Quantum Optimization with Continuous Bandits , author=. arXiv preprint arXiv:2502.04021 , year=

    work page arXiv

  3. [3]

    Advances in neural information processing systems , volume=

    Reinforcement learning for optimization of variational quantum circuit architectures , author=. Advances in neural information processing systems , volume=

  4. [4]

    Zur Elektrodynamik bewegter Körper

    Albert Einstein. Zur Elektrodynamik bewegter Körper. Annalen der Physik. 1905

    1905

  5. [5]

    The \ Companion

    Michel Goossens and Frank Mittelbach and Alexander Samarin. The \ Companion. 1993

    1993

  6. [6]

    CoRR , volume =

    Neural contextual bandits with deep representation and shallow exploration , author=. arXiv preprint arXiv:2012.01780 , year=

    work page arXiv 2012

  7. [7]

    Quantum Machine Intelligence , volume=

    On fundamental aspects of quantum extreme learning machines , author=. Quantum Machine Intelligence , volume=. 2025 , publisher=

    2025

  8. [8]

    Quantum Science and Technology , volume=

    Quantum-classical hybrid neural networks in the neural tangent kernel regime , author=. Quantum Science and Technology , volume=. 2023 , publisher=

    2023

  9. [9]

    Quantum , volume=

    Quantum lazy training , author=. Quantum , volume=. 2023 , publisher=

    2023

  10. [10]

    International Conference on Machine Learning , pages=

    Neural contextual bandits with ucb-based exploration , author=. International Conference on Machine Learning , pages=. 2020 , organization=

    2020

  11. [11]

    Trained quantum neural networks are Gaussian processes

    Trained quantum neural networks are Gaussian processes , author=. arXiv preprint arXiv:2402.08726 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  12. [12]

    New Journal of Physics , volume=

    Improving the dynamics of quantum sensors with reinforcement learning , author=. New Journal of Physics , volume=. 2020 , publisher=

    2020

  13. [13]

    Physical Review X , volume=

    Reinforcement learning in different phases of quantum control , author=. Physical Review X , volume=. 2018 , publisher=

    2018

  14. [14]

    Physical review letters , volume=

    Quantum support vector machine for big data classification , author=. Physical review letters , volume=. 2014 , publisher=

    2014

  15. [15]

    npj Quantum Information , volume=

    Quantum classifier with tailored quantum kernel , author=. npj Quantum Information , volume=. 2020 , publisher=

    2020

  16. [16]

    Quantum Machine Intelligence , volume=

    Quantum Gaussian process regression for Bayesian optimization , author=. Quantum Machine Intelligence , volume=. 2024 , publisher=

    2024

  17. [17]

    Machine Learning: Science and Technology , volume=

    Benchmarking of quantum fidelity kernels for Gaussian process regression , author=. Machine Learning: Science and Technology , volume=. 2024 , publisher=

    2024

  18. [18]

    International Conference on Machine Learning , pages=

    Learning near optimal policies with low inherent bellman error , author=. International Conference on Machine Learning , pages=. 2020 , organization=

    2020

  19. [19]

    Conference on Learning Theory , pages=

    Efficient and robust algorithms for adversarial linear contextual bandits , author=. Conference on Learning Theory , pages=. 2020 , organization=

    2020

  20. [20]

    International Conference on Machine Learning , pages=

    High-dimensional experimental design and kernel bandits , author=. International Conference on Machine Learning , pages=. 2021 , organization=

    2021

  21. [21]

    Proceedings of the AAAI Conference on Artificial Intelligence , volume=

    Misspecified linear bandits , author=. Proceedings of the AAAI Conference on Artificial Intelligence , volume=

  22. [22]

    arXiv preprint arXiv:2301.09680 , year=

    Quantum heavy-tailed bandits , author=. arXiv preprint arXiv:2301.09680 , year=

    work page arXiv

  23. [23]

    Advances in Neural Information Processing Systems , volume=

    Quantum bayesian optimization , author=. Advances in Neural Information Processing Systems , volume=

  24. [24]

    2017 , publisher=

    Fourier analysis on groups , author=. 2017 , publisher=

    2017

  25. [25]

    The 40th Conference on Uncertainty in Artificial Intelligence , year=

    Quantum kernelized bandits , author=. The 40th Conference on Uncertainty in Artificial Intelligence , year=

  26. [26]

    arXiv preprint arXiv:2311.13552 , year=

    A unified framework for trace-induced quantum kernels , author=. arXiv preprint arXiv:2311.13552 , year=

    work page arXiv

  27. [27]

    2023 , isbn =

    Wan, Zongqi and Zhang, Zhijie and Li, Tongyang and Zhang, Jialin and Sun, Xiaoming , title =. 2023 , isbn =. doi:10.1609/aaai.v37i8.26202 , booktitle =

    work page doi:10.1609/aaai.v37i8.26202 2023

  28. [28]

    Quantum , volume=

    Multi-armed quantum bandits: Exploration versus exploitation when learning properties of quantum states , author=. Quantum , volume=. 2022 , publisher=

    2022

  29. [29]

    Advances in Neural Information Processing Systems , volume=

    The inductive bias of quantum kernels , author=. Advances in Neural Information Processing Systems , volume=

  30. [30]

    In International Conference on Machine Learning (ICML), pages 1015--1022 , year=

    Gaussian process optimization in the bandit setting: No regret and experimental design , author=. In International Conference on Machine Learning (ICML), pages 1015--1022 , year=

  31. [31]

    Nature Communications , volume=

    Universal expressiveness of variational quantum classifiers and quantum kernels for support vector machines , author=. Nature Communications , volume=. 2023 , publisher=

    2023

  32. [32]

    Quantum Science and Technology , volume=

    Filtering variational quantum algorithms for combinatorial optimization , author=. Quantum Science and Technology , volume=. 2022 , publisher=

    2022

  33. [33]

    International Conference on Machine Learning , pages=

    Approximation theory based methods for rkhs bandits , author=. International Conference on Machine Learning , pages=. 2021 , organization=

    2021

  34. [34]

    Finite-Time Analysis of Kernelised Contextual Bandits

    Finite-time analysis of kernelised contextual bandits , author=. arXiv preprint arXiv:1309.6869 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  35. [35]

    arXiv preprint arXiv:2112.00426 , year=

    Noisy Bayesian optimization for variational quantum eigensolvers , author=. arXiv preprint arXiv:2112.00426 , year=

    work page arXiv

  36. [36]

    npj Quantum Information , volume=

    Variational quantum algorithm with information sharing , author=. npj Quantum Information , volume=. 2021 , publisher=

    2021

  37. [37]

    Science advances , volume=

    Training of quantum circuits on a hybrid quantum computer , author=. Science advances , volume=. 2019 , publisher=

    2019

  38. [38]

    Advances in neural information processing systems , volume=

    Misspecified gaussian process bandit optimization , author=. Advances in neural information processing systems , volume=

  39. [39]

    Advances in Neural Information Processing Systems , volume=

    Physics-informed bayesian optimization of variational quantum circuits , author=. Advances in Neural Information Processing Systems , volume=

  40. [40]

    Gaussian Processes and Kernel Methods: A Review on Connections and Equivalences

    Gaussian processes and kernel methods: A review on connections and equivalences , author=. arXiv preprint arXiv:1807.02582 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  41. [41]

    Effect of data encoding on the expressive power of variational quantum-machine-learning models , volume=

    Schuld, Maria and Sweke, Ryan and Meyer, Johannes Jakob , year=. Effect of data encoding on the expressive power of variational quantum-machine-learning models , volume=. Physical Review A , publisher=. doi:10.1103/physreva.103.032430 , number=

    work page doi:10.1103/physreva.103.032430

  42. [42]

    Machine Learning: Science and Technology , volume=

    On the expressivity of embedding quantum kernels , author=. Machine Learning: Science and Technology , volume=. 2024 , publisher=

    2024

  43. [43]

    arXiv preprint arXiv:2409.04406 , year=

    Quantum Kernel Methods under Scrutiny: A Benchmarking Study , author=. arXiv preprint arXiv:2409.04406 , year=

    work page arXiv

  44. [44]

    Gaussian Process Optimization in the Bandit Setting: No Regret and Experimental Design

    Gaussian process optimization in the bandit setting: No regret and experimental design , author=. arXiv preprint arXiv:0912.3995 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  45. [45]

    Nature communications , volume=

    Power of data in quantum machine learning , author=. Nature communications , volume=. 2021 , publisher=

    2021

  46. [46]

    Quantum Science and Technology , volume=

    Faster variational quantum algorithms with quantum kernel-based surrogate models , author=. Quantum Science and Technology , volume=. 2023 , publisher=

    2023

  47. [47]

    Machine Learning with Quantum Computers , pages=

    Quantum models as kernel methods , author=. Machine Learning with Quantum Computers , pages=. 2021 , publisher=

    2021

  48. [48]

    Nature communications , volume=

    Noise-induced barren plateaus in variational quantum algorithms , author=. Nature communications , volume=. 2021 , publisher=

    2021

  49. [49]

    Nature communications , volume=

    Barren plateaus in quantum neural network training landscapes , author=. Nature communications , volume=. 2018 , publisher=

    2018

  50. [50]

    Nature Reviews Physics , volume=

    Variational quantum algorithms , author=. Nature Reviews Physics , volume=. 2021 , publisher=

    2021

  51. [51]

    Nature communications , volume=

    Exponential concentration in quantum kernel methods , author=. Nature communications , volume=. 2024 , publisher=

    2024

  52. [52]

    Quantum Machine Intelligence , volume=

    Quantum bandits , author=. Quantum Machine Intelligence , volume=. 2020 , publisher=

    2020

  53. [53]

    Proceedings of the AAAI Conference on Artificial Intelligence , volume=

    Quantum exploration algorithms for multi-armed bandits , author=. Proceedings of the AAAI Conference on Artificial Intelligence , volume=

  54. [54]

    , title =

    Li, Lihong and Chu, Wei and Langford, John and Schapire, Robert E. , title =. 2010 , booktitle =

    2010

  55. [55]

    2017 , booktitle =

    Ghosh, Avishek and Chowdhury, Sayak Ray and Gopalan, Aditya , title =. 2017 , booktitle =

    2017

  56. [56]

    Learning with good feature representations in bandits and in RL with a generative model , year =

    Lattimore, Tor and Szepesv\'. Learning with good feature representations in bandits and in RL with a generative model , year =. Proceedings of the 37th International Conference on Machine Learning , articleno =

  57. [57]

    and Rakhlin, Alexander , title =

    Foster, Dylan J. and Rakhlin, Alexander , title =. Proceedings of the 37th International Conference on Machine Learning , articleno =. 2020 , publisher =

    2020

  58. [58]

    and Xu, Yuesheng and Zhang, Haizhang , title =

    Micchelli, Charles A. and Xu, Yuesheng and Zhang, Haizhang , title =. J. Mach. Learn. Res. , month = dec, pages =. 2006 , issue_date =

    2006

  59. [59]

    Proceedings of the 21st International Conference on Neural Information Processing Systems , pages =

    Rahimi, Ali and Recht, Benjamin , title =. Proceedings of the 21st International Conference on Neural Information Processing Systems , pages =. 2007 , publisher =

    2007

  60. [60]

    2022 , eprint=

    Classically Approximating Variational Quantum Machine Learning with Random Fourier Features , author=. 2022 , eprint=

    2022

  61. [61]

    2021 , eprint=

    Supervised quantum machine learning models are kernel methods , author=. 2021 , eprint=

    2021

  62. [62]

    2015 , eprint=

    On the Error of Random Fourier Features , author=. 2015 , eprint=

    2015

  63. [63]

    Advances in Computational Mathematics , volume=

    Near-optimal data-independent point locations for radial basis function interpolation , author=. Advances in Computational Mathematics , volume=. 2005 , publisher=

    2005

  64. [64]

    Constructive Approximation , pages=

    Greedy Algorithms for Reduced Bases in Banach Spaces , author=. Constructive Approximation , pages=. 2013 , publisher=

    2013

  65. [65]

    and Huang, Hsin-Yuan and Cerezo, M

    Caro, Matthias C. and Huang, Hsin-Yuan and Cerezo, M. and Sharma, Kunal and Sornborger, Andrew and Cincio, Lukasz and Coles, Patrick J. , year=. Generalization in quantum machine learning from few training data , volume=. Nature Communications , publisher=. doi:10.1038/s41467-022-32550-3 , number=

    work page doi:10.1038/s41467-022-32550-3

  66. [66]

    Kolmogorov widths on the sphere via eigenvalue estimates for H\"{o}lderian integral operators

    Jordão, T. and Menegatto, V.A. , year=. Kolmogorov widths on the sphere via eigenvalue estimates for H\". 1805.08476 , archivePrefix=

    work page internal anchor Pith review Pith/arXiv arXiv

  67. [67]

    A Course in Functional Analysis , author =

  68. [68]

    Quantum Machine Learning in Feature Hilbert Spaces , volume=

    Schuld, Maria and Killoran, Nathan , year=. Quantum Machine Learning in Feature Hilbert Spaces , volume=. Physical Review Letters , publisher=. doi:10.1103/physrevlett.122.040504 , number=

    work page doi:10.1103/physrevlett.122.040504

  69. [69]

    Supervised learning with quantum-enhanced feature spaces,

    Havlicek, Vojtech and Corcoles, Antonio D. and Temme, Kristan and Harrow, Aram W. and Kandala, Abhinav and Chow, Jerry M. and Gambetta, Jay M. , year=. Supervised learning with quantum-enhanced feature spaces , volume=. Nature , publisher=. doi:10.1038/s41586-019-0980-2 , number=

    work page doi:10.1038/s41586-019-0980-2

  70. [70]

    On the sub-Gaussianity of the Beta and Dirichlet distributions , volume=

    Marchal, Olivier and Arbel, Julyan , year=. On the sub-Gaussianity of the Beta and Dirichlet distributions , volume=. Electronic Communications in Probability , publisher=. doi:10.1214/17-ecp92 , number=

    work page doi:10.1214/17-ecp92

  71. [71]

    Uniform approximation of functions with random bases , year=

    Rahimi, Ali and Recht, Benjamin , booktitle=. Uniform approximation of functions with random bases , year=

  72. [72]

    doi:10.22331/q-2024-11-14-1523 , url =

    Gradients and frequency profiles of quantum re-uploading models , author =. doi:10.22331/q-2024-11-14-1523 , url =

    work page doi:10.22331/q-2024-11-14-1523 2024

  73. [73]

    2018 , eprint=

    What Doubling Tricks Can and Can't Do for Multi-Armed Bandits , author=. 2018 , eprint=

    2018

  74. [74]

    and Towsley, Don , title =

    Wang, Xuchuang and Chen, Yu-Zhen Janice and de Andrade, Matheus Guedes and Allcock, Jonathan and Hajiesmaili, Mohammad and Lui, John C.S. and Towsley, Don , title =. 2025 , isbn =. doi:10.1609/aaai.v39i20.35432 , booktitle =

    work page doi:10.1609/aaai.v39i20.35432 2025

  75. [75]

    Buchholz, Simon and K. Multi-. doi:10.22331/q-2025-03-25-1672 , url =

    work page doi:10.22331/q-2025-03-25-1672 2025

  76. [76]

    Wang, Daochen and You, Xuchen and Li, Tongyang and Childs, Andrew M. , year=. Quantum Exploration Algorithms for Multi-Armed Bandits , volume=. Proceedings of the AAAI Conference on Artificial Intelligence , publisher=. doi:10.1609/aaai.v35i11.17212 , number=

    work page doi:10.1609/aaai.v35i11.17212