Quantum-Inspired Robust and Scalable SAR Object Classification

Benjamin Rodrigues de Miranda; Felix Bock; Marco Trenti; Maximilian Scharf; Padraig Davidson; Sigurd Huber; Timo Felser; Tobias Brosch

arxiv: 2604.25755 · v2 · pith:GR4O5OSHnew · submitted 2026-04-28 · 🪐 quant-ph · cs.CV· physics.comp-ph

Quantum-Inspired Robust and Scalable SAR Object Classification

Maximilian Scharf , Marco Trenti , Felix Bock , Padraig Davidson , Tobias Brosch , Benjamin Rodrigues de Miranda , Sigurd Huber , Timo Felser This is my paper

Pith reviewed 2026-05-25 06:31 UTC · model grok-4.3

classification 🪐 quant-ph cs.CVphysics.comp-ph

keywords tensor networksSAR classificationdata poisoning robustnessmodel compressionradar object detectionquantum-inspired methodsedge device deployment

0 comments

The pith

Tensor networks deliver robust SAR object classification that resists data poisoning while shrinking model size for edge devices.

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

The paper sets out to show that tensor networks can classify objects in SAR images despite heavy noise and wide dynamic range, while also resisting data poisoning attacks and requiring fewer parameters than standard neural networks. This combination matters for deployment on drones and aircraft where both reliability under attack and limited compute are constraints. The work shifts focus from conventional neural networks to tensor networks specifically for their robustness and compression properties in radar settings. If correct, it suggests tensor networks can meet the dual demands of accuracy under corruption and efficiency in real radar pipelines.

Core claim

Tensor networks achieve both resilience to data poisoning and model compression in SAR object classification tasks, outperforming conventional neural networks in these respects under the evaluated conditions.

What carries the argument

Tensor networks applied to SAR image classification, performing the dual work of maintaining classification accuracy under noise and data poisoning while reducing parameter count for edge deployment.

If this is right

SAR classification models can be made smaller without sacrificing robustness to noise or poisoning.
Deployment on resource-limited radar platforms becomes feasible with maintained reliability.
Tensor networks supply a route to efficient deep learning methods beyond conventional neural networks for radar tasks.

Where Pith is reading between the lines

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

The same tensor-network approach might extend to other high-noise sensor modalities such as sonar or medical ultrasound.
If the compression holds across datasets, it could reduce the energy cost of real-time radar processing on aircraft.
The robustness finding invites direct comparison of tensor networks against adversarial training techniques on identical SAR benchmarks.

Load-bearing premise

The tested SAR datasets and poisoning scenarios actually reveal superior robustness and scalability for tensor networks compared with conventional neural networks.

What would settle it

A follow-up experiment on the same SAR data in which tensor networks exhibit equal or lower robustness to poisoning attacks or require equal or larger models than neural networks to reach target accuracy.

read the original abstract

SAR image classification naturally has to deal with huge noise and a high dynamic range particularly requiring robust classification models. Additionally, the deployment of these models on edge devices, such as drones and military aircraft, requires a careful balance between model size and classification accuracy. This study explores the potential of tensor networks to meet these robustness requirements, specifically evaluating their resilience to data poisoning. Unlike previous works that concentrated on conventional neural networks for SAR object detection, this research focuses on the robustness and model reduction capabilities of tensor networks in object classification. Our findings indicate that tensor networks are adept at addressing both the challenges of robustness and the need for model efficiency, thereby contributing valuable insights to the ongoing discourse in radar applications and deep learning methodologies in general.

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.

Desk Editor's Note private letter to a colleague

The paper applies tensor networks to SAR classification and claims gains in poisoning robustness plus efficiency, but the abstract supplies no experiments, metrics, or comparisons to back it up.

read the letter

The core of this work is taking tensor networks, which are already used in machine learning for compression and some robustness properties, and testing them on SAR object classification with an eye toward data poisoning attacks and running on edge hardware like drones. That matches a practical need in radar where noise and dynamic range are high and model size matters. The abstract positions this as different from prior neural net work on SAR detection by emphasizing robustness and reduction. That focus is reasonable and the domain application is clear enough. What is actually new appears limited to the specific use case rather than any new tensor network technique or theoretical result. The paper does well in naming the dual requirements of robustness and efficiency up front. The soft spots are more substantial. The abstract states findings on robustness without describing the dataset, the poisoning method, the attack strength, the baseline models and their sizes, the tensor network architecture, or any accuracy numbers or drop metrics. The stress-test concern lands directly: without those details it is impossible to tell whether any reported advantage comes from the tensor network structure or from differences in capacity or training. No equations or derivations are shown either, so there is nothing to check for circularity or soundness. This leaves the central claim unsupported on the evidence given. The work is for people working on applied ML for radar or sensor data who want to see tensor networks tried in that setting. A reader already familiar with tensor networks in ML will not find a new framework here. It deserves a serious referee if the full paper contains reproducible experiments and fair baselines, because the problem is real even if the current presentation is thin. I would send it to review rather than desk reject, but only to get the actual results on the table.

Referee Report

1 major / 0 minor

Summary. The manuscript investigates tensor networks as a quantum-inspired approach for SAR object classification, emphasizing their potential for robustness against data poisoning and improved scalability/model efficiency suitable for edge deployment (e.g., drones), in contrast to prior work focused on conventional neural networks.

Significance. If the claimed experimental advantages in robustness and efficiency are substantiated with proper controls and baselines, the work could contribute a novel inductive bias for handling noisy, high-dynamic-range SAR data while reducing model size, with relevance to both radar applications and tensor-network methods in ML.

major comments (1)

[Abstract] Abstract: The central claim that 'tensor networks are adept at addressing both the challenges of robustness and the need for model efficiency' (and specifically superior resilience to data poisoning) rests on 'our findings,' yet the text provides no description of the SAR dataset, poisoning protocol (attack type, fraction, label vs. feature), tensor-network architecture/contraction, baseline CNN parameter counts, training procedure, or quantitative metrics (accuracy drop, etc.). This absence makes it impossible to evaluate whether any advantage is due to the tensor-network bias or to uncontrolled differences in capacity/regularization.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their feedback. We address the concern regarding the abstract below.

read point-by-point responses

Referee: [Abstract] Abstract: The central claim that 'tensor networks are adept at addressing both the challenges of robustness and the need for model efficiency' (and specifically superior resilience to data poisoning) rests on 'our findings,' yet the text provides no description of the SAR dataset, poisoning protocol (attack type, fraction, label vs. feature), tensor-network architecture/contraction, baseline CNN parameter counts, training procedure, or quantitative metrics (accuracy drop, etc.). This absence makes it impossible to evaluate whether any advantage is due to the tensor-network bias or to uncontrolled differences in capacity/regularization.

Authors: The abstract is written as a concise overview; the full manuscript details the MSTAR SAR dataset, the label-flipping poisoning protocol (including attack fractions and type), the MPS tensor-network architecture and contraction scheme, baseline CNN parameter counts, training procedures, and quantitative accuracy metrics in Sections 2–4. We agree that the abstract would benefit from a brief mention of these elements to allow immediate evaluation of the claims. We will revise the abstract to incorporate key specifics such as the dataset, poisoning method, and example metrics while remaining within length constraints. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical claims rest on experiments, not derivations

full rationale

The manuscript is an empirical comparison of tensor networks versus conventional neural networks on SAR classification tasks, with claims about robustness to data poisoning and scalability supported by experimental findings. No derivation chain, equations, or first-principles results are described in the provided abstract or skeptic summary. The central assertions are therefore not reducible by construction to inputs via self-definition, fitted parameters renamed as predictions, or self-citation load-bearing steps. This is the expected non-finding for an experimental methods paper without mathematical derivations.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review; no free parameters, axioms, or invented entities can be identified from the provided text.

pith-pipeline@v0.9.0 · 5669 in / 921 out tokens · 27768 ms · 2026-05-25T06:31:32.218600+00:00 · methodology

Review history (2 revisions) →

discussion (0)

Reference graph

Works this paper leans on

34 extracted references · 34 canonical work pages

[1]

Deep Big Simple Neural Nets for Handwritten Digit Recogntion

D. C. Ciresan et al. “Deep Big Simple Neural Nets for Handwritten Digit Recogntion”. In:Neural Computation22.12 (2010), pp. 3207–20

work page 2010
[2]

Offline Handwrit- ing Recognition withMultidimensional Recurrent Neural Networks

A. Graves and J. Schmidhuber. “Offline Handwrit- ing Recognition withMultidimensional Recurrent Neural Networks”. In:Advances in Neural Infor- mation Processing Systems 21. NIPS. Curran Asso- ciates, Inc., 2009, pp. 545–552

work page 2009
[3]

Ima- geNet Classification with Deep Convolutional Neu- ral Networks

A. Krizhevsky, I. Sutskever, and G. E. Hinton. “Ima- geNet Classification with Deep Convolutional Neu- ral Networks”. In:NIPS. 2012

work page 2012
[4]

Backpropagation Applied to Hand- written Zip Code Recognition

Y . LeCun et al. “Backpropagation Applied to Hand- written Zip Code Recognition”. In:Neural Compu- tation1.4 (1989), pp. 541–51

work page 1989
[5]

Deep Learning Ap- proach for Radar Applications

C. Neumann and T. Brosch. “Deep Learning Ap- proach for Radar Applications”. In:21st Interna- tional Radar Symposium (IRS). 2020, pp. 28–30

work page 2020
[6]

Micro-Doppler based detection and tracking of UA Vs with multistatic radar

F. Hoffmann et al. “Micro-Doppler based detection and tracking of UA Vs with multistatic radar”. In: 2016 IEEE Radar Conference (RadarConf). 2016, pp. 1–6

work page 2016
[7]

Micro-Doppler Based Target Recog- nition With Radars: A Review

A. Hanif et al. “Micro-Doppler Based Target Recog- nition With Radars: A Review”. In:IEEE Sensors Journal22.4 (2022), pp. 2948–2961

work page 2022
[8]

A Realtime Micro-Doppler De- tection, Tracking and Classification System for the 94 GHz FMCW Radar System DUSIM

R. Petervari et al. “A Realtime Micro-Doppler De- tection, Tracking and Classification System for the 94 GHz FMCW Radar System DUSIM”. In:2019 16th European Radar Conference (EuRAD). 2019, pp. 193–196

work page 2019
[9]

Micro-Doppler based target classification in ground surveillance radar sys- tems

R. Amiri and A. Shahzadi. “Micro-Doppler based target classification in ground surveillance radar sys- tems”. In:Digital Signal Processing101 (2020), p. 102702

work page 2020
[10]

Automatic Target Recognition on High Resolution SAR

T. Brosch and C. Neumann. “Automatic Target Recognition on High Resolution SAR”. In:21st In- ternational Radar Symposium (IRS). IEEE, 2021, pp. 1–6

work page 2021
[11]

Ultrafast Object Detection on High Resolution SAR

M. Schwaiger et al. “Ultrafast Object Detection on High Resolution SAR”. In:International Radar Symposium. 2022, pp. 1–5

work page 2022
[12]

Almalioglu et al.SAR Object Detection with Self-Supervised Pretraining and Curriculum-Aware Sampling

Y . Almalioglu et al.SAR Object Detection with Self-Supervised Pretraining and Curriculum-Aware Sampling. 2025

work page 2025
[13]

Object Detection on SAR Im- ages Via YOLOv10 and Integrated ACmix Attention Mechanism

Z. Wang et al. “Object Detection on SAR Im- ages Via YOLOv10 and Integrated ACmix Attention Mechanism”. In:2024 6th International Conference on Robotics, Intelligent Control and Artificial Intel- ligence (RICAI). 2024, pp. 756–760

work page 2024
[14]

Maritime Object Detection in Synthetic Aperture Radar (SAR) Imagery Using Deep Learning

M. Warner et al. “Maritime Object Detection in Synthetic Aperture Radar (SAR) Imagery Using Deep Learning”. In:AGU Fall Meeting Abstracts. V ol. 2020. AGU Fall Meeting Abstracts. Dec. 2020, G004-0022, G004–0022

work page 2020
[15]

Generalization in deep learning- based aircraft classification for SAR imagery

A. Pulella et al. “Generalization in deep learning- based aircraft classification for SAR imagery”. In: ISPRS Journal of Photogrammetry and Remote Sensing218 (2024), pp. 312–323

work page 2024
[16]

MSTAR Extended Operating Conditions: A Tutorial

E. R. Keydel, S. W. Lee, and J. T. Moore. “MSTAR Extended Operating Conditions: A Tutorial”. In: vol. 2757. International Society for Optics and Pho- tonics. SPIE, 1996, pp. 228–42

work page 1996
[17]

Standard SAR ATR Evaluation Ex- periments Using the MSTAR Public Release Data Set

T. Ross et al. “Standard SAR ATR Evaluation Ex- periments Using the MSTAR Public Release Data Set”. In:Algorithms for Synthetic Aperture Radar Imagery V. V ol. 3370. SPIE, 1998, pp. 566–73

work page 1998
[18]

A Review of Synthetic–Aperture Radar Image Formation Algorithms and Implemen- tations A Computational Perspective

H. Cruz et al. “A Review of Synthetic–Aperture Radar Image Formation Algorithms and Implemen- tations A Computational Perspective”. In:Remote Sensing14.1258 (2022), pp. 1–30

work page 2022
[19]

The density-matrix renormaliza- tion group in the age of matrix product states

U. Schollwöck. “The density-matrix renormaliza- tion group in the age of matrix product states”. In: Annals of Physics326.1 (Jan. 2011), 96–192

work page 2011
[20]

Montangero.Introduction to Tensor Network Methods

S. Montangero.Introduction to Tensor Network Methods. Springer International Publishing, 2018

work page 2018
[21]

Optimizing radiotherapy plans for cancer treatment with Tensor Networks

S. Cavinato et al. “Optimizing radiotherapy plans for cancer treatment with Tensor Networks”. In: Physics in Medicine & Biology66.12 (June 2021), p. 125015

work page 2021
[22]

E. M. Stoudenmire and D. J. Schwab.Supervised Learning with Quantum-Inspired Tensor Networks. 2017

work page 2017
[23]

Quantum-inspired machine learning on high-energy physics data

T. Felser et al. “Quantum-inspired machine learning on high-energy physics data”. In:npj Quantum In- formation. 1st ser. 7 (2021)

work page 2021
[24]

Solving gapped Hamiltonians lo- cally

M. B. Hastings. “Solving gapped Hamiltonians lo- cally”. In:Physical Review B73.8 (Feb. 2006)

work page 2006
[25]

Algorithms for entangle- ment renormalization

G. Evenbly and G. Vidal. “Algorithms for entangle- ment renormalization”. In:Physical Review B79.14 (Apr. 2009)

work page 2009
[26]

Effi- cient Tensor Network Ansatz for High-Dimensional Quantum Many-Body Problems

T. Felser, S. Notarnicola, and S. Montangero. “Effi- cient Tensor Network Ansatz for High-Dimensional Quantum Many-Body Problems”. In:Physical Re- view Letters126.17 (Apr. 2021)

work page 2021
[27]

Criticality, the Area Law, and the Computational Power of Projected Entangled Pair States

F. Verstraete et al. “Criticality, the Area Law, and the Computational Power of Projected Entangled Pair States”. In:Physical Review Letters96.22 (June 2006)

work page 2006
[28]

Willner, M

M. Willner, M. Trenti, and D. Lebiedz.Riemannian Optimization on Tree Tensor Networks with Appli- cation in Machine Learning. 2025

work page 2025
[29]

Glasser, N

I. Glasser, N. Pancotti, and J. I. Cirac.From proba- bilistic graphical models to generalized tensor net- works for supervised learning. 2019

work page 2019
[30]

Learning relevant features of data with multi-scale tensor networks

E. M. Stoudenmire. “Learning relevant features of data with multi-scale tensor networks”. In:Quantum Science and Technology3.3 (Apr. 2018), p. 034003

work page 2018
[31]

Review of Recent Advances in AI/ML Using the MSTAR Data

E. Blasch et al. “Review of Recent Advances in AI/ML Using the MSTAR Data”. In:Algo- rithms for Synthetic Aperture Radar Imagery XXVII. V ol. 11393. SPIE, 2020, p. 113930C

work page 2020
[32]

Fein-Ashley et al.Benchmarking Deep Learning Classifiers for SAR Automatic Target Recognition

J. Fein-Ashley et al.Benchmarking Deep Learning Classifiers for SAR Automatic Target Recognition. 2023

work page 2023
[33]

ATR of Battle- field Targets by SAR – Classification Results Us- ing the Public MSTAR Dataset Compared with a Dataset by QinetiQ, UK

R. Schumacher and K. Rosenbach. “ATR of Battle- field Targets by SAR – Classification Results Us- ing the Public MSTAR Dataset Compared with a Dataset by QinetiQ, UK”. In:Research and Technol- ogy Organization, Sensors and Electronics Technol- ogy Panel (Symposium). RTO–MP–SET–080, 2004, pp. 31–1–31–12

work page 2004
[34]

Singh and V

A. Singh and V . K. Singh.Exploring Deep Learning Methods for Classification of SAR Images: Towards NextGen Convolutions via Transformers. 2023

work page 2023

[1] [1]

Deep Big Simple Neural Nets for Handwritten Digit Recogntion

D. C. Ciresan et al. “Deep Big Simple Neural Nets for Handwritten Digit Recogntion”. In:Neural Computation22.12 (2010), pp. 3207–20

work page 2010

[2] [2]

Offline Handwrit- ing Recognition withMultidimensional Recurrent Neural Networks

A. Graves and J. Schmidhuber. “Offline Handwrit- ing Recognition withMultidimensional Recurrent Neural Networks”. In:Advances in Neural Infor- mation Processing Systems 21. NIPS. Curran Asso- ciates, Inc., 2009, pp. 545–552

work page 2009

[3] [3]

Ima- geNet Classification with Deep Convolutional Neu- ral Networks

A. Krizhevsky, I. Sutskever, and G. E. Hinton. “Ima- geNet Classification with Deep Convolutional Neu- ral Networks”. In:NIPS. 2012

work page 2012

[4] [4]

Backpropagation Applied to Hand- written Zip Code Recognition

Y . LeCun et al. “Backpropagation Applied to Hand- written Zip Code Recognition”. In:Neural Compu- tation1.4 (1989), pp. 541–51

work page 1989

[5] [5]

Deep Learning Ap- proach for Radar Applications

C. Neumann and T. Brosch. “Deep Learning Ap- proach for Radar Applications”. In:21st Interna- tional Radar Symposium (IRS). 2020, pp. 28–30

work page 2020

[6] [6]

Micro-Doppler based detection and tracking of UA Vs with multistatic radar

F. Hoffmann et al. “Micro-Doppler based detection and tracking of UA Vs with multistatic radar”. In: 2016 IEEE Radar Conference (RadarConf). 2016, pp. 1–6

work page 2016

[7] [7]

Micro-Doppler Based Target Recog- nition With Radars: A Review

A. Hanif et al. “Micro-Doppler Based Target Recog- nition With Radars: A Review”. In:IEEE Sensors Journal22.4 (2022), pp. 2948–2961

work page 2022

[8] [8]

A Realtime Micro-Doppler De- tection, Tracking and Classification System for the 94 GHz FMCW Radar System DUSIM

R. Petervari et al. “A Realtime Micro-Doppler De- tection, Tracking and Classification System for the 94 GHz FMCW Radar System DUSIM”. In:2019 16th European Radar Conference (EuRAD). 2019, pp. 193–196

work page 2019

[9] [9]

Micro-Doppler based target classification in ground surveillance radar sys- tems

R. Amiri and A. Shahzadi. “Micro-Doppler based target classification in ground surveillance radar sys- tems”. In:Digital Signal Processing101 (2020), p. 102702

work page 2020

[10] [10]

Automatic Target Recognition on High Resolution SAR

T. Brosch and C. Neumann. “Automatic Target Recognition on High Resolution SAR”. In:21st In- ternational Radar Symposium (IRS). IEEE, 2021, pp. 1–6

work page 2021

[11] [11]

Ultrafast Object Detection on High Resolution SAR

M. Schwaiger et al. “Ultrafast Object Detection on High Resolution SAR”. In:International Radar Symposium. 2022, pp. 1–5

work page 2022

[12] [12]

Almalioglu et al.SAR Object Detection with Self-Supervised Pretraining and Curriculum-Aware Sampling

Y . Almalioglu et al.SAR Object Detection with Self-Supervised Pretraining and Curriculum-Aware Sampling. 2025

work page 2025

[13] [13]

Object Detection on SAR Im- ages Via YOLOv10 and Integrated ACmix Attention Mechanism

Z. Wang et al. “Object Detection on SAR Im- ages Via YOLOv10 and Integrated ACmix Attention Mechanism”. In:2024 6th International Conference on Robotics, Intelligent Control and Artificial Intel- ligence (RICAI). 2024, pp. 756–760

work page 2024

[14] [14]

Maritime Object Detection in Synthetic Aperture Radar (SAR) Imagery Using Deep Learning

M. Warner et al. “Maritime Object Detection in Synthetic Aperture Radar (SAR) Imagery Using Deep Learning”. In:AGU Fall Meeting Abstracts. V ol. 2020. AGU Fall Meeting Abstracts. Dec. 2020, G004-0022, G004–0022

work page 2020

[15] [15]

Generalization in deep learning- based aircraft classification for SAR imagery

A. Pulella et al. “Generalization in deep learning- based aircraft classification for SAR imagery”. In: ISPRS Journal of Photogrammetry and Remote Sensing218 (2024), pp. 312–323

work page 2024

[16] [16]

MSTAR Extended Operating Conditions: A Tutorial

E. R. Keydel, S. W. Lee, and J. T. Moore. “MSTAR Extended Operating Conditions: A Tutorial”. In: vol. 2757. International Society for Optics and Pho- tonics. SPIE, 1996, pp. 228–42

work page 1996

[17] [17]

Standard SAR ATR Evaluation Ex- periments Using the MSTAR Public Release Data Set

T. Ross et al. “Standard SAR ATR Evaluation Ex- periments Using the MSTAR Public Release Data Set”. In:Algorithms for Synthetic Aperture Radar Imagery V. V ol. 3370. SPIE, 1998, pp. 566–73

work page 1998

[18] [18]

A Review of Synthetic–Aperture Radar Image Formation Algorithms and Implemen- tations A Computational Perspective

H. Cruz et al. “A Review of Synthetic–Aperture Radar Image Formation Algorithms and Implemen- tations A Computational Perspective”. In:Remote Sensing14.1258 (2022), pp. 1–30

work page 2022

[19] [19]

The density-matrix renormaliza- tion group in the age of matrix product states

U. Schollwöck. “The density-matrix renormaliza- tion group in the age of matrix product states”. In: Annals of Physics326.1 (Jan. 2011), 96–192

work page 2011

[20] [20]

Montangero.Introduction to Tensor Network Methods

S. Montangero.Introduction to Tensor Network Methods. Springer International Publishing, 2018

work page 2018

[21] [21]

Optimizing radiotherapy plans for cancer treatment with Tensor Networks

S. Cavinato et al. “Optimizing radiotherapy plans for cancer treatment with Tensor Networks”. In: Physics in Medicine & Biology66.12 (June 2021), p. 125015

work page 2021

[22] [22]

E. M. Stoudenmire and D. J. Schwab.Supervised Learning with Quantum-Inspired Tensor Networks. 2017

work page 2017

[23] [23]

Quantum-inspired machine learning on high-energy physics data

T. Felser et al. “Quantum-inspired machine learning on high-energy physics data”. In:npj Quantum In- formation. 1st ser. 7 (2021)

work page 2021

[24] [24]

Solving gapped Hamiltonians lo- cally

M. B. Hastings. “Solving gapped Hamiltonians lo- cally”. In:Physical Review B73.8 (Feb. 2006)

work page 2006

[25] [25]

Algorithms for entangle- ment renormalization

G. Evenbly and G. Vidal. “Algorithms for entangle- ment renormalization”. In:Physical Review B79.14 (Apr. 2009)

work page 2009

[26] [26]

Effi- cient Tensor Network Ansatz for High-Dimensional Quantum Many-Body Problems

T. Felser, S. Notarnicola, and S. Montangero. “Effi- cient Tensor Network Ansatz for High-Dimensional Quantum Many-Body Problems”. In:Physical Re- view Letters126.17 (Apr. 2021)

work page 2021

[27] [27]

Criticality, the Area Law, and the Computational Power of Projected Entangled Pair States

F. Verstraete et al. “Criticality, the Area Law, and the Computational Power of Projected Entangled Pair States”. In:Physical Review Letters96.22 (June 2006)

work page 2006

[28] [28]

Willner, M

M. Willner, M. Trenti, and D. Lebiedz.Riemannian Optimization on Tree Tensor Networks with Appli- cation in Machine Learning. 2025

work page 2025

[29] [29]

Glasser, N

I. Glasser, N. Pancotti, and J. I. Cirac.From proba- bilistic graphical models to generalized tensor net- works for supervised learning. 2019

work page 2019

[30] [30]

Learning relevant features of data with multi-scale tensor networks

E. M. Stoudenmire. “Learning relevant features of data with multi-scale tensor networks”. In:Quantum Science and Technology3.3 (Apr. 2018), p. 034003

work page 2018

[31] [31]

Review of Recent Advances in AI/ML Using the MSTAR Data

E. Blasch et al. “Review of Recent Advances in AI/ML Using the MSTAR Data”. In:Algo- rithms for Synthetic Aperture Radar Imagery XXVII. V ol. 11393. SPIE, 2020, p. 113930C

work page 2020

[32] [32]

Fein-Ashley et al.Benchmarking Deep Learning Classifiers for SAR Automatic Target Recognition

J. Fein-Ashley et al.Benchmarking Deep Learning Classifiers for SAR Automatic Target Recognition. 2023

work page 2023

[33] [33]

ATR of Battle- field Targets by SAR – Classification Results Us- ing the Public MSTAR Dataset Compared with a Dataset by QinetiQ, UK

R. Schumacher and K. Rosenbach. “ATR of Battle- field Targets by SAR – Classification Results Us- ing the Public MSTAR Dataset Compared with a Dataset by QinetiQ, UK”. In:Research and Technol- ogy Organization, Sensors and Electronics Technol- ogy Panel (Symposium). RTO–MP–SET–080, 2004, pp. 31–1–31–12

work page 2004

[34] [34]

Singh and V

A. Singh and V . K. Singh.Exploring Deep Learning Methods for Classification of SAR Images: Towards NextGen Convolutions via Transformers. 2023

work page 2023