pith. sign in

arxiv: 2512.11367 · v2 · submitted 2025-12-12 · 🪐 quant-ph · cs.LG

Maritime object classification with SAR imagery using quantum kernel methods

John Tanner , Nicholas Davies , Pascal Jahan Elahi , Casey R. Myers , Du Huynh , Wei Liu , Mark Reynolds , Jingbo Wang This is my paper

Pith reviewed 2026-05-16 23:23 UTC · model grok-4.3

classification 🪐 quant-ph cs.LG
keywords quantum kernel methodsSAR imagerymaritime classificationvessel detectionquantum machine learningsynthetic aperture radarIUU fishing
0
0 comments X

The pith

Quantum kernel methods achieve equal or better performance than classical kernels when classifying maritime objects in real SAR imagery.

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

The paper investigates whether quantum kernel methods can improve classification of small maritime objects like vessels and fishing boats in synthetic aperture radar images. SAR provides all-weather surveillance crucial for combating illegal fishing, which causes major economic losses. By testing quantum kernels on real and complex SAR chips from the SARFish dataset against classical kernels, the work shows that in noiseless simulations quantum approaches can match or surpass classical performance on real data. However, quantum encoding of complex data leads to overfitting. This marks the first application of quantum kernel methods to this maritime task.

Core claim

In noiseless numerical simulations, quantum kernel methods applied to real SAR chips obtain equal or better classification performance than classical Laplacian, radial basis function, and linear kernels for distinguishing vessels from non-vessels and fishing vessels from other vessels, while the quantum kernel on complex SAR data overfits and underperforms.

What carries the argument

Quantum kernel methods that map SAR image chips to quantum states and compute similarity via quantum measurements.

If this is right

  • Quantum kernels provide a competitive tool for SAR-based maritime object classification.
  • Real-valued SAR chips are better suited for quantum encoding than complex ones in this setup.
  • The approach demonstrates potential for quantum machine learning in remote sensing applications.
  • Restricting comparisons to kernel-based models ensures fairness when evaluating quantum versus classical performance.

Where Pith is reading between the lines

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

  • Future quantum hardware with reduced noise could enable practical deployment of these kernels for real-time maritime monitoring.
  • Similar quantum methods might apply to other SAR classification tasks beyond vessels, such as oil spill or ice detection.
  • Developing quantum kernels more robust to complex-valued inputs could expand their utility across signal processing domains.

Load-bearing premise

The assumption that noiseless simulation results will translate to useful performance on actual quantum hardware despite noise and that dataset splits do not bias the comparisons.

What would settle it

Implementing the quantum kernel methods on noisy intermediate-scale quantum hardware and finding that their accuracy drops below that of classical kernels on the same SAR dataset would falsify the observed advantage.

Figures

Figures reproduced from arXiv: 2512.11367 by Casey R. Myers, Du Huynh, Jingbo Wang, John Tanner, Mark Reynolds, Nicholas Davies, Pascal Jahan Elahi, Wei Liu.

Figure 1
Figure 1. Figure 1: FIG. 1 [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2 [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4 [PITH_FULL_IMAGE:figures/full_fig_p012_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5 [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗
read the original abstract

Illegal, unreported, and unregulated (IUU) fishing causes global economic losses of 10-25 billion USD annually and undermines marine sustainability and governance. Synthetic Aperture Radar (SAR) provides reliable maritime surveillance under all weather and lighting conditions, but classifying small maritime objects in SAR imagery remains challenging. We investigate quantum machine learning for this task, focusing on quantum kernel methods (QKMs) applied to real and complex SAR chips extracted from the SARFish dataset. We tackle two binary classification problems, the first for distinguishing vessels from non-vessels, and the second for distinguishing fishing vessels from other types of vessels. We compare QKMs applied to real and complex SAR chips against classical Laplacian, RBF, and linear kernels applied to real SAR chips. We restrict the comparison to be between just kernel based models so that the comparison is as fair and meaningful as possible. Using noiseless numerical simulations of the quantum kernels, we find that with the real SAR chips, QKMs are capable of obtaining equal or better performance than the classical kernels in the best case. However, the specific quantum kernel used to encode the complex SAR data overfits and performs poorly. This work presents the first application of QKMs to maritime classification in SAR imagery and offers insight into the potential and current limitations of quantum-enhanced learning for maritime surveillance.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 2 minor

Summary. The manuscript applies quantum kernel methods (QKMs) to two binary classification tasks on real and complex SAR chips from the SARFish dataset: vessel versus non-vessel and fishing vessel versus other vessels. It compares noiseless simulations of QKMs against classical Laplacian, RBF, and linear kernels (restricted to real SAR chips) and reports that QKMs achieve equal or better accuracy in the best case on real data while the complex-data encoding overfits.

Significance. If the noiseless results prove robust, the work would constitute the first application of QKMs to SAR maritime surveillance and supply concrete empirical evidence that quantum kernels can at least match classical kernel performance on this domain. The explicit restriction to kernel-based models and the acknowledgment of overfitting on complex encodings are positive features that keep the comparison focused.

major comments (2)
  1. [Abstract and simulation results] Abstract and simulation results: the headline claim that QKMs are 'capable of obtaining equal or better performance' rests entirely on noiseless numerical evaluation of the Gram matrices. Because quantum kernel entries are estimated from circuit samples, any realistic noise model (depolarizing, readout, or decoherence) will perturb those entries; the manuscript contains no noisy simulations or hardware runs, leaving open whether the observed margin survives on actual devices.
  2. [Results section] Results section: no error bars, standard deviations across random seeds, or statistical significance tests (e.g., paired t-tests or Wilcoxon tests) are reported for the accuracy comparisons. Without these, it is impossible to judge whether the 'equal or better' outcome is statistically distinguishable from classical kernels or simply within simulation variance.
minor comments (2)
  1. [Abstract] The abstract states 'in the best case' without identifying which quantum kernel or hyper-parameter setting achieves the reported performance; a brief qualifier would improve clarity.
  2. [Methods] Dataset preprocessing and train/test split details are referenced but not fully specified (e.g., exact chip sizes, normalization, or stratification); adding a short table or paragraph would aid reproducibility.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed comments. We address each major point below and indicate the revisions we will implement.

read point-by-point responses

  1. Referee: [Abstract and simulation results] Abstract and simulation results: the headline claim that QKMs are 'capable of obtaining equal or better performance' rests entirely on noiseless numerical evaluation of the Gram matrices. Because quantum kernel entries are estimated from circuit samples, any realistic noise model (depolarizing, readout, or decoherence) will perturb those entries; the manuscript contains no noisy simulations or hardware runs, leaving open whether the observed margin survives on actual devices.

    Authors: We agree that the reported results are obtained from noiseless simulations and that this constitutes a limitation for claims about practical quantum hardware performance. In the revised manuscript we will add a new subsection presenting simulations that incorporate standard noise models (depolarizing channels and readout errors) applied to the quantum kernel circuits. We will also expand the discussion to clarify the distinction between the noiseless results and expected behavior on near-term devices. revision: yes

  2. Referee: [Results section] Results section: no error bars, standard deviations across random seeds, or statistical significance tests (e.g., paired t-tests or Wilcoxon tests) are reported for the accuracy comparisons. Without these, it is impossible to judge whether the 'equal or better' outcome is statistically distinguishable from classical kernels or simply within simulation variance.

    Authors: We acknowledge that the current results lack error bars and statistical tests, which prevents a rigorous assessment of whether performance differences are significant. We will recompute all accuracy figures over multiple random seeds, add error bars showing standard deviations, and include paired t-tests (or Wilcoxon signed-rank tests where appropriate) comparing quantum and classical kernels. These statistical results and updated figures will appear in the revised results section. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical head-to-head kernel comparison

full rationale

The paper reports direct numerical results from noiseless simulations of quantum kernels versus classical Laplacian/RBF/linear kernels on fixed SARFish dataset splits. No equation or claim reduces a reported accuracy or F1 score to a fitted parameter defined inside the paper, nor invokes a self-citation chain to establish uniqueness of the chosen encoding. The comparison is explicitly restricted to kernel methods for fairness, and performance numbers are obtained by standard training and testing rather than by any self-definitional or renaming step. The analysis is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The work relies on standard quantum kernel constructions and classical kernel SVMs with no new axioms or invented entities introduced; any kernel hyperparameters are implicit in the ML setup but not enumerated as free parameters in the abstract.

pith-pipeline@v0.9.0 · 5551 in / 1031 out tokens · 29255 ms · 2026-05-16T23:23:41.233455+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

79 extracted references · 79 canonical work pages

  1. [1]

    El-Darymli, P

    K. El-Darymli, P. McGuire, D. Power, and C. Moloney. Target detection in synthetic aperture radar imagery: a state-of-the-art survey.Journal of Applied Remote Sensing, 7(1):071598, March 2013

    work page 2013

  2. [2]

    J. Li, Y. Liu, X. Wang, Z. Jiang, and Y. Li. A robust cfar algorithm based on superpixel merging operation 16 for sar ship detection. InProceedings of the 2024 7th International Conference on Image and Graphics Pro- cessing, ICIGP ’24, page 388–393, New York, NY, USA,

    work page 2024

  3. [3]

    Association for Computing Machinery

    work page

  4. [4]

    X. Wen, S. Zhang, J. Wang, T. Yao, and Y. Tang. A cfar-enhanced ship detector for sar images based on yolov5s.Remote Sensing, 16(5), 2024

    work page 2024

  5. [5]

    C. Wang, B. Guo, Ji. Song, F. He, and C. Li. A novel cfar-based ship detection method using range- compressed data for spaceborne sar system.IEEE Transactions on Geoscience and Remote Sensing, 62:1– 15, 2024

    work page 2024

  6. [6]

    Færch, W

    L. Færch, W. Dierking, N. Hughes, and A. P. Doul- geris. A comparison of constant false alarm rate object detection algorithms for iceberg identification in l- and c-band sar imagery of the labrador sea.The Cryosphere, 17(12):5335–5355, 2023

    work page 2023

  7. [7]

    T. Xie, M. Liu, M. Zhang, S. Qi, and J. Yang. Ship detection based on a superpixel-level cfar detector for sar imagery.International Journal of Remote Sensing, 43(9):3412–3428, 2022

    work page 2022

  8. [8]

    Schwegmann, W

    C. Schwegmann, W. Kleynhans, B. P. Salmon, L. Mdakane, and R. Meyer. Very deep learning for ship discrimination in synthetic aperture radar imagery. pages 104–107, 07 2016

    work page 2016

  9. [9]

    Zhang, X

    T. Zhang, X. Zhang, X. Ke, X. Zhan, J. Shi, S. Wei, D. Pan, J. Li, H. Su, Y. Zhou, and D. Kumar. Ls-ssdd- v1.0: A deep learning dataset dedicated to small ship detection from large-scale sentinel-1 sar images.Remote Sensing, 12(18), 2020

    work page 2020

  10. [10]

    J. Li, C. Xu, H. Su, L. Gao, and T. Wang. Deep learning for sar ship detection: Past, present and future.Remote Sensing, 14(11), 2022

    work page 2022

  11. [11]

    Gupta, O

    H. Gupta, O. P. Verma, T. K. Sharma, H. Varshney, S. Agarwal, and W. Pak. Ship detection using ensemble deep learning techniques from synthetic aperture radar imagery.Scientific Reports, 14(1):29397, 2024

    work page 2024

  12. [12]

    Redmon, S

    J. Redmon, S. Divvala, R. Girshick, and A. Farhadi. You only look once: Unified, real-time object detection, 2016

    work page 2016

  13. [13]

    Yasir, S

    M. Yasir, S. Liu, S. Pirasteh, M. Xu, H. Sheng, J. Wan, F. A. P. de Figueiredo, F. J. Aguilar, and J. Li. Yoloshiptracker: Tracking ships in sar images using lightweight yolov8.International Journal of Applied Earth Observation and Geoinformation, 134:104137, 2024

    work page 2024

  14. [14]

    H. Wang, J. Shi, H. Karimian, F. Liu, and F. Wang. Yolosar-lite: a lightweight framework for real-time ship detection in sar imagery.International Journal of Dig- ital Earth, 17(1):2405525, 2024

    work page 2024

  15. [15]

    Selvam, P

    P. Selvam, P. S. Sundari, M. Tamilselvi, T. Suresh, M. Murugappan, and M. E. H. Chowdhury. Yolo- sail: Attention-enhanced yolov5 with optimized bi-fpn for ship target detection in sar images.IEEE access, 13:29523–29540, 2025

    work page 2025

  16. [16]

    Williams, B

    J. Williams, B. Yip, B. McCarthy, T.-T. Cao, and A. Robles-Kelly. Contrastive learning for ship classi- fication using real and complex sar imagery. In2023 International Conference on Digital Image Computing: Techniques and Applications (DICTA), pages 531–538, 2023

    work page 2023

  17. [17]

    C. Wang, H. Zhang, F. Wu, B. Zhang, and S. Tian. Ship classification with deep learning using cosmo-skymed sar data. In2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), pages 558–561, 2017

    work page 2017

  18. [18]

    Y. Guan, X. Zhang, S. Chen, G. Liu, Y. Jia, Y. Zhang, G. Gao, J. Zhang, Z. Li, and C. Cao. Fishing vessel classification in sar images using a novel deep learning model.IEEE Transactions on Geoscience and Remote Sensing, 61:1–21, 2023

    work page 2023

  19. [19]

    Y. Wang, C. Wang, and H. Zhang. Ship classification in high-resolution sar images using deep learning of small datasets.Sensors, 18(9), 2018

    work page 2018

  20. [20]

    A. A. Al Hinai and R. Guida. Confidence-aware ship classification using contour features in sar images.Re- mote Sensing, 17(1), 2025

    work page 2025

  21. [21]

    Lang and S

    H. Lang and S. Wu. Ship classification in moderate- resolution sar image by naive geometric features- combined multiple kernel learning.IEEE Geoscience and Remote Sensing Letters, PP:1–5, 09 2017

    work page 2017

  22. [22]

    H. Lang, J. Zhang, X. Zhang, and J. Meng. Ship clas- sification in sar image by joint feature and classifier se- lection.IEEE Geoscience and Remote Sensing Letters, 13(2):212–216, 2016

    work page 2016

  23. [23]

    K. Ji, X. Xing, W. Chen, H. Zou, and J. Chen. Ship classification in terrasar-x sar images based on classifier combination. In2013 IEEE International Geoscience and Remote Sensing Symposium - IGARSS, pages 2589– 2592, 2013

    work page 2013

  24. [24]

    Z. Yan, X. Song, L. Yang, and Y. Wang. Ship classifica- tion in synthetic aperture radar images based on multi- ple classifiers ensemble learning and automatic identifi- cation system data transfer learning.Remote Sensing, 14(21), 2022

    work page 2022

  25. [25]

    Okafor-Yarwood

    I. Okafor-Yarwood. Illegal, unreported and unregulated fishing, and the complexities of the sustainable devel- opment goals (sdgs) for countries in the gulf of guinea. Marine Policy, 99:414–422, 2019

    work page 2019

  26. [26]

    D. J. Agnew, J. Pearce, G. Pramod, T. Peatman, R. Watson, J. R. Beddington, and T. J. Pitcher. Es- timating the worldwide extent of illegal fishing.PLOS ONE, 4(2):1–8, 02 2009

    work page 2009

  27. [27]

    Welch, T

    H. Welch, T. Clavelle, T. D. White, M. A. Cimino, J. V. Osdel, T. Hochberg, D. Kroodsma, and E. L. Hazen. Hot spots of unseen fishing vessels.Science Advances, 17 8(44):eabq2109, 2022

    work page 2022

  28. [28]

    Desai and G

    R. Desai and G. Shambaugh. Measuring the global impact of destructive and illegal fishing on maritime piracy: A spatial analysis.PLOS ONE, 16:e0246835, 02 2021

    work page 2021

  29. [29]

    Torres, P

    R. Torres, P. Snoeij, D. Geudtner, D. Bibby, M. David- son, E. Attema, P. Potin, B. Rommen, N. Floury, M. Brown, I. Navas Traver, P. Deghaye, B. Dues- mann, B. Rosich, N. Miranda, C. Bruno, M. L’Abbate, R. Croci, A. Pietropaolo, M. Huchler, and F. Rostan. Gmes sentinel-1 mission.Remote Sensing of Environ- ment, 120:9–24, 2012. The Sentinel Missions - New ...

    work page 2012

  30. [30]

    Luckett, B

    C. Luckett, B. McCarthy, T.-T. Cao, and A. Robles- Kelly. The sarfish dataset and challenge. InProceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 752–761, 2024

    work page 2024

  31. [31]

    Schuld and F

    M. Schuld and F. Petruccione.Supervised Learning with Quantum Computers. Springer International Publish- ing, 2018

    work page 2018

  32. [32]

    Schuld and F

    M. Schuld and F. Petruccione.Quantum Models as Kernel Methods, pages 217–245. Springer International Publishing, 2021. DOI: 10.1007/978-3-030-83098-4 6

    work page doi:10.1007/978-3-030-83098-4 2021

  33. [33]

    Nadim, M

    M. Nadim, M. Hassan, A. K. Mandal, C. K. Roy, B. Roy, and K. A. Schneider. Comparative analysis of quantum and classical support vector classifiers for software bug prediction: an exploratory study.Quantum Machine Intelligence, 7(1), March 2025

    work page 2025

  34. [34]

    Havl´ ıˇ cek, A

    V. Havl´ ıˇ cek, A. D. C´ orcoles, K. Temme, A. W. Harrow, A. Kandala, J. M. Chow, and J. M. Gambetta. Super- vised learning with quantum-enhanced feature spaces. Nature 567, 209–212 (2019)

    work page 2019

  35. [35]

    Y. Liu, S. Arunachalam, and K. Temme. A rigorous and robust quantum speed-up in supervised machine learning. Nat. Phys. 17, 1013–1017 (2021)

    work page 2021

  36. [36]

    Huang, M

    H.-Y. Huang, M. Broughton, M. Mohseni, R. Babbush, S. Boixo, H. Neven, and J. R. McClean. Power of data in quantum machine learning. Nat. Commun. 12, 2631 (2021)

    work page 2021

  37. [37]

    Y. Wu, B. Wu, J. Wang, and X. Yuan. Quantum phase recognition via quantum kernel methods. Quantum 7, 981 (2023)

    work page 2023

  38. [38]

    Giovagnoli, S

    A. Giovagnoli, S. Huber, and G. Krieger. A quantum range-doppler algorithm for synthetic aperture radar image formation, 2025

    work page 2025

  39. [39]

    Al Salahat, M

    K. Al Salahat, M. El Moussawi, and A. J. Ghandour. Quantum meets sar: A novel range-doppler algorithm for next-gen earth observation, 2025

    work page 2025

  40. [40]

    Mauro, A

    F. Mauro, A. Sebastianelli, M. P. Del Rosso, P. Gamba, and S. L. Ullo. Qspecklefilter: a quantum machine learning approach for sar speckle filtering, 2024

    work page 2024

  41. [41]

    L. Wang, Y. Liu, F. Meng, T. Luan, W. Liu, Z. Zhang, and X. Yu. A quantum synthetic aperture radar im- age denoising algorithm based on grayscale morphology. iScience, 27(5):109627–, 2024

    work page 2024

  42. [42]

    Miller, G

    L. Miller, G. Uehara, and A. Spanias. Quantum im- age fusion methods for remote sensing. In2024 IEEE Aerospace Conference, pages 1–9, 2024

    work page 2024

  43. [43]

    Miller, G

    L. Miller, G. Uehara, A. Sharma, and A. Spanias. Quan- tum machine learning for optical and sar classification. In2023 24th International Conference on Digital Signal Processing (DSP), pages 1–5, 2023

    work page 2023

  44. [44]

    S. Naik, N. Vaughn, G. Uehara, A. Spanias, and K. Jaskie. Quantum classification for synthetic aperture radar. InAutomatic Target Recognition XXXIV, Pro- ceedings of SPIE - The International Society for Optical Engineering. SPIE, 2024

    work page 2024

  45. [45]

    T.-T. V. Cao, C. Luckett, J. Williams, T. Cooke, B. Yip, A. Rajagopalan, and S. Wong. Sarfish: Space-based maritime surveillance using complex synthetic aperture radar imagery.2022 International Conference on Dig- ital Image Computing: Techniques and Applications (DICTA), pages 1–8, 2022

    work page 2022

  46. [46]

    Paolo, T

    F. Paolo, T. T. Lin, R. Gupta, B. Goodman, N. Patel, D. Kuster, D. Kroodsma, and J. Dunnmon. xview3-sar: Detecting dark fishing activity using synthetic aperture radar imagery, 2022

    work page 2022

  47. [47]

    Simonyan and A

    K. Simonyan and A. Zisserman. Very deep convolutional networks for large-scale image recognition, 2015

    work page 2015

  48. [48]

    F. Chollet. Xception: Deep learning with depthwise sep- arable convolutions. In2017 IEEE Conference on Com- puter Vision and Pattern Recognition (CVPR), pages 1800–1807, 2017

    work page 2017

  49. [49]

    Szegedy, V

    C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna. Rethinking the inception architecture for computer vision. In2016 IEEE Conference on Com- puter Vision and Pattern Recognition (CVPR), pages 2818–2826, 2016

    work page 2016

  50. [50]

    Schmitt, L

    M. Schmitt, L. H. Hughes, and X. X. Zhu. The sen1- 2 dataset for deep learning in sar-optical data fusion, 2018

    work page 2018

  51. [51]

    Air Force

    Sandia National Laboratories U.S. Air Force. Mstar public dataset.https://www.sdms.afrl.af.mil/ index.php?collection=mstar, 1995. Accessed via the Sensor Data Management System (SDMS), X-band spotlight SAR, 10 target types

    work page 1995

  52. [52]

    Nielsen and I

    M. Nielsen and I. Chuang.Quantum Computation and Quantum Information. Cambridge University Press, 2000

    work page 2000

  53. [53]

    Sch¨ olkopf and A

    B. Sch¨ olkopf and A. J. Smola.Learning with Ker- nels: Support Vector Machines, Regularization, Opti- mization, and Beyond. MIT Press, Cambridge, MA, USA, 2001

    work page 2001

  54. [54]

    Mohri, A

    M. Mohri, A. Rostamizadeh, and A. Talwalkar.Foun- dations of Machine Learning. MIT Press, Cambridge, MA, USA, 2018. 18

    work page 2018

  55. [55]

    J. Platt. Sequential minimal optimization: A fast algo- rithm for training support vector machines. Technical Report MSR-TR-98-14, Microsoft, April 1998

    work page 1998

  56. [56]

    H. W. Kuhn and A. W. Tucker.Nonlinear Program- ming, pages 247–258. Springer Basel, Basel, 2014

    work page 2014

  57. [57]

    Karush.Minima of Functions of Several Variables with Inequalities as Side Conditions, pages 217–245

    W. Karush.Minima of Functions of Several Variables with Inequalities as Side Conditions, pages 217–245. Springer Basel, Basel, 2014

    work page 2014

  58. [58]

    Rifkin, G

    R. Rifkin, G. Yeo, and T. Poggio. Regularized least- squares classification.Advances in Learning Theory: Methods, Model and Applications, NATO Science Se- ries III: Computer and Systems Sciences, 190, 06 2003

    work page 2003

  59. [59]

    Fung and O

    G. Fung and O. L. Mangasarian. Proximal support vec- tor machine classifiers. KDD ’01, page 77–86, New York, NY, USA, 2001. Association for Computing Machinery

    work page 2001

  60. [60]

    F. Le Gall. Powers of tensors and fast matrix multipli- cation. InProceedings of the 39th International Sympo- sium on Symbolic and Algebraic Computation, ISSAC ’14, page 296–303, New York, NY, USA, 2014. Associ- ation for Computing Machinery

    work page 2014

  61. [61]

    A. E. Paine, V. E. Elfving, and O. Kyriienko. Quan- tum kernel methods for solving regression problems and differential equations.Phys. Rev. A, 107:032428, Mar 2023

    work page 2023

  62. [62]

    Schnabel and M

    J. Schnabel and M. Roth. Quantum kernel methods un- der scrutiny: a benchmarking study.Quantum Machine Intelligence, 7(1), apr 2025

    work page 2025

  63. [63]

    Larocca, S

    M. Larocca, S. Thanasilp, S. Wang, K. Sharma, J. Bia- monte, P. J. Coles, L. Cincio, J. R. McClean, Z. Holmes, and M. Cerezo. Barren plateaus in variational quan- tum computing.Nature Reviews Physics, 7(4):174–189, March 2025

    work page 2025

  64. [64]

    Jerbi, L

    S. Jerbi, L. J. Fiderer, H. P. Nautrup, J. M. K¨ ubler, H. J. Briegel, and V. Dunjko. Quantum machine learn- ing beyond kernel methods. Nat Commun 14, 517 (2023)

    work page 2023

  65. [65]

    J. M. K¨ ubler, S. Buchholz, and B. Sch¨ olkopf. The induc- tive bias of quantum kernels. 2021. arXiv:2106.03747

    work page arXiv 2021

  66. [66]

    Thanasilp, S

    S. Thanasilp, S. Wang, Cerezo M., and Z. Holmes. Exponential concentration in quantum kernel methods. Nat Commun 15, 5200 (2024)

    work page 2024

  67. [67]

    Hubregtsen, D

    T. Hubregtsen, D. Wierichs, E. Gil-Fuster, P.-J. H. S. Derks, P. K. Faehrmann, and J. J. Meyer. Train- ing quantum embedding kernels on near-term quantum computers. Phys. Rev. A 106, 042431 (2022)

    work page 2022

  68. [68]

    C. R. Harris, K. J. Millman, S. J. van der Walt, R. Gom- mers, P. Virtanen, D. Cournapeau, E. Wieser, J. Tay- lor, S. Berg, N. J. Smith, R. Kern, M. Picus, S. Hoyer, M. H. van Kerkwijk, M. Brett, A. Haldane, J. F. del R´ ıo, M. Wiebe, P. Peterson, P. G´ erard-Marchant, K. Shep- pard, T. Reddy, W. Weckesser, H. Abbasi, C. Gohlke, and T. E. Oliphant. Array ...

    work page 2020

  69. [69]

    Canatar, E

    A. Canatar, E. Peters, C. Pehlevan, S. M. Wild, and R. Shaydulin. Bandwidth enables generalization in quantum kernel models, 2023

    work page 2023

  70. [70]

    Shaydulin and S

    R. Shaydulin and S. M. Wild. Importance of kernel bandwidth in quantum machine learning.Phys. Rev. A, 106:042407, Oct 2022

    work page 2022

  71. [71]

    Tanner, J

    J. Tanner, J. Pye, and J. Wang. Learning out-of-time- ordered correlators with classical kernel methods.Phys. Rev. B, 111:144301, Apr 2025

    work page 2025

  72. [72]

    Virmani and H

    D. Virmani and H. Pandey. Comparative analysis on effect of different svm kernel functions for classifica- tion. In Deepak Gupta, Ashish Khanna, Aboul Ella Hassanien, Sameer Anand, and Ajay Jaiswal, edi- tors,International Conference on Innovative Comput- ing and Communications, pages 657–670, Singapore,

    work page

  73. [73]

    Springer Nature Singapore

    work page

  74. [74]

    Yekkehkhany, A

    B. Yekkehkhany, A. Safari, S. Homayouni, and M. Hasanlou. A comparison study of different kernel functions for svm-based classification of multi-temporal polarimetry sar data.ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial In- formation Sciences, XL-2/W3:281–285, 10 2014

    work page 2014

  75. [75]

    Kandala, A

    A. Kandala, A. Mezzacapo, K. Temme, M. Takita, M. Brink, J. M. Chow, and J. M. Gambetta. Hardware-efficient variational quantum eigensolver for small molecules and quantum magnets.Nature, 549(7671):242–246, 2017

    work page 2017

  76. [76]

    Leone, S

    L. Leone, S. F. E. Oliviero, L. Cincio, and M. Cerezo. On the practical usefulness of the hardware efficient ansatz.Quantum, 8:1395, July 2024

    work page 2024

  77. [77]

    T. Haug, C. N. Self, and M. S. Kim. Quantum ma- chine learning of large datasets using randomized mea- surements.Machine Learning: Science and Technology, 4(1):015005, jan 2023

    work page 2023

  78. [78]

    Pedregosa, G

    F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, and V. Dubourg et al. Scikit-learn: Machine learning in Python. JMLR 12(85), 2825-2830 (2011)

    work page 2011

  79. [79]

    Javadi-Abhari, M

    A. Javadi-Abhari, M. Treinish, K. Krsulich, C. J. Wood, J. Lishman, J. Gacon, S. Martiel, P. D. Nation, L. S. Bishop, A. W. Cross, B. R. Johnson, and J. M. Gam- betta. Quantum computing with Qiskit, 2024. 19 Appendix A: Hyperparameter values In this appendix, Table II provides a list of the hyperparameter values which were trialled for each kernel during ...

    work page 2024