pith. sign in

arxiv: 2003.00768 · v2 · submitted 2020-03-02 · 📡 eess.SP · cs.CV· cs.LG

Convolutional Sparse Support Estimator Network (CSEN) From energy efficient support estimation to learning-aided Compressive Sensing

Mehmet Yamac , Mete Ahishali , Serkan Kiranyaz , Moncef Gabbouj This is my paper

Pith reviewed 2026-05-24 15:00 UTC · model grok-4.3

classification 📡 eess.SP cs.CVcs.LG
keywords support estimationcompressive sensingconvolutional neural networkssparse signalsmachine learningedge devicessignal recovery
0
0 comments X

The pith

A compact convolutional network learns to map compressive measurements directly to the support locations of a sparse signal.

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

The paper introduces Convolutional Support Estimator Networks to learn a mapping from compressive measurements to the positions of non-zero elements in sparse signals. Traditional approaches typically reconstruct the entire signal first using iterative methods before identifying the support. By training on pairs of measurements and supports, CSEN achieves comparable accuracy to state-of-the-art methods but with much lower computational demands. This makes it practical for real-time applications on energy-constrained devices and allows its output to serve as prior information that can enhance conventional sparse recovery techniques.

Core claim

Convolutional Support Estimator Networks (CSENs) are compact convolutional architectures designed to directly estimate the support set from compressively sensed measurements by learning from training data consisting of measurement-support pairs. This bypasses the conventional pipeline of first recovering the sparse signal via greedy or optimization methods and then extracting the support. The networks can operate in real-time on low-power devices for tasks such as anomaly localization and can provide prior information to improve the performance of sparse signal recovery algorithms, all while attaining state-of-the-art performance levels with significantly reduced computational complexity.

What carries the argument

The Convolutional Support Estimator Network (CSEN), a compact convolutional neural network that learns to predict the indices of non-zero coefficients directly from the compressive measurements.

If this is right

  • Support estimation can be performed in real time on mobile and low-power edge devices.
  • The output of CSEN can be used as prior information to improve sparse signal recovery algorithms.
  • State-of-the-art performance is achieved with significantly lower computational complexity than iterative methods.
  • Applications include anomaly localization and simultaneous face recognition.

Where Pith is reading between the lines

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

  • If the training data covers a wide range of sparsity levels and signal types, CSEN could reduce the need for problem-specific tuning in compressive sensing systems.
  • Hybrid approaches combining CSEN with traditional recovery might yield even better results than either alone.
  • This direct mapping approach could extend to other inverse problems where estimating discrete parameters from measurements is key.

Load-bearing premise

A representative training set of measurement-support pairs must exist such that the learned mapping generalizes to unseen signals without retraining.

What would settle it

Evaluating the CSEN on test signals drawn from a different distribution than the training set, such as signals with varying sparsity ratios, and checking whether its support estimation accuracy falls below that of standard iterative algorithms like OMP or Basis Pursuit.

Figures

Figures reproduced from arXiv: 2003.00768 by Mehmet Yamac, Mete Ahishali, Moncef Gabbouj, Serkan Kiranyaz.

Figure 1
Figure 1. Figure 1: The proposed CSEN with two potential applications: a) (bottom-left) Sparse Support Estimation b) (top-middle) [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Most common model for a practical support estimator. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Proposed model for an efficient support estimator. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Type-I Convolutional Support Estimator Network (CSEN1). [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Type-II Convolutional Support Estimator Network (CSEN2). [PITH_FULL_IMAGE:figures/full_fig_p006_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Histogram of ρi’s obtained from the 10k samples (test set). The vectorized gray scale images, xi in MNIST dataset are already sparse in the spatial domain (in canonical basis, i.e., Φ = I) with kxik ≤ ki . The experiments in this study have been carried out on a workstation that has four Nvidia R TITAN-X GPU cards and Intel R Xeon(R) CPU E5-2637 v4 at 3.50GHz with 128 GB memory. Tensorflow library [55] is … view at source ↗
Figure 7
Figure 7. Figure 7: F1 Measure graph of CSEN and Lamp configurations [PITH_FULL_IMAGE:figures/full_fig_p007_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Reconstruction accuracy vs. process time comparison [PITH_FULL_IMAGE:figures/full_fig_p007_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Reconstruction accuracy vs. process time comparison [PITH_FULL_IMAGE:figures/full_fig_p008_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: The graphical representation of proposed dictionary [PITH_FULL_IMAGE:figures/full_fig_p009_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: It is clear that the proposed approach recovers the [PITH_FULL_IMAGE:figures/full_fig_p009_11.png] view at source ↗
Figure 11
Figure 11. Figure 11: Examples from MNIST that are compressively sensed, and then reconstructed at MR= [PITH_FULL_IMAGE:figures/full_fig_p010_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: (Top) Proposed Compressive Sensing Reconstruction. [PITH_FULL_IMAGE:figures/full_fig_p010_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Phase Transition of the Algorithms. (MP) [67], Orthogonal Matching Pursuit (OMP) [8], Compres￾sive Sampling Matched Pursuit (CoSaMP) [9]. iii) Bayesian Framework [68], etc. These conventional algorithms dealing with sparse inverse problems work in an iterative manner; for instance, most convex relaxation techniques such as BPDN, minimize the data fidelity term (e.g., `2-norm) and sparsifying term (e.g., `… view at source ↗
read the original abstract

Support estimation (SE) of a sparse signal refers to finding the location indices of the non-zero elements in a sparse representation. Most of the traditional approaches dealing with SE problem are iterative algorithms based on greedy methods or optimization techniques. Indeed, a vast majority of them use sparse signal recovery techniques to obtain support sets instead of directly mapping the non-zero locations from denser measurements (e.g., Compressively Sensed Measurements). This study proposes a novel approach for learning such a mapping from a training set. To accomplish this objective, the Convolutional Support Estimator Networks (CSENs), each with a compact configuration, are designed. The proposed CSEN can be a crucial tool for the following scenarios: (i) Real-time and low-cost support estimation can be applied in any mobile and low-power edge device for anomaly localization, simultaneous face recognition, etc. (ii) CSEN's output can directly be used as "prior information" which improves the performance of sparse signal recovery algorithms. The results over the benchmark datasets show that state-of-the-art performance levels can be achieved by the proposed approach with a significantly reduced computational complexity.

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 paper proposes Convolutional Sparse Support Estimator Networks (CSEN), compact CNNs that learn a direct mapping from compressive measurements to the support set of a sparse signal. The approach targets real-time low-power edge applications (e.g., anomaly detection) and supplies a prior that can accelerate or improve traditional sparse-recovery algorithms. Experiments on benchmark datasets are reported to reach state-of-the-art support-estimation accuracy while incurring substantially lower computational cost than iterative greedy or optimization baselines.

Significance. If the performance and complexity claims are substantiated by rigorous, reproducible experiments, the work would offer a practical route to low-complexity support estimation and a new class of learned priors for compressive sensing. The explicit framing for edge-device deployment and the potential to reduce iteration counts in downstream recovery algorithms are concrete strengths.

major comments (2)
  1. [Experiments] Experiments section (and associated tables/figures): the manuscript asserts SOTA performance and complexity reduction yet supplies no quantitative error metrics (e.g., support-recovery F1, Hamming distance), baseline algorithms with identical measurement matrices and sparsity levels, or wall-clock / FLOPs comparisons. Without these data the central claim cannot be evaluated.
  2. [Method and Experiments] Training / test protocol (method and experiments sections): no description is given of how the training measurement-support pairs are generated relative to the test distribution (sparsity level, noise statistics, signal class). The generalization assumption required for the fixed CSEN to deliver the claimed accuracy on unseen signals is therefore untested; a domain-shift experiment or cross-validation across signal classes is needed.
minor comments (2)
  1. [Introduction / Method] Notation for the measurement matrix and support vector should be introduced once and used consistently; several symbols appear without prior definition.
  2. [Method] Figure captions for network diagrams should explicitly state input/output dimensions and the precise loss used during training.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We address each major point below and will revise the manuscript accordingly to strengthen the experimental evaluation and clarify the data-generation protocol.

read point-by-point responses

  1. Referee: [Experiments] Experiments section (and associated tables/figures): the manuscript asserts SOTA performance and complexity reduction yet supplies no quantitative error metrics (e.g., support-recovery F1, Hamming distance), baseline algorithms with identical measurement matrices and sparsity levels, or wall-clock / FLOPs comparisons. Without these data the central claim cannot be evaluated.

    Authors: We agree that the current presentation of results would benefit from additional quantitative metrics. The revised manuscript will report support-recovery F1 scores and Hamming distances, include baseline comparisons that use identical measurement matrices and sparsity levels, and provide FLOPs counts together with wall-clock timings on the same hardware platform. These additions will be placed in the experiments section and associated tables/figures. revision: yes

  2. Referee: [Method and Experiments] Training / test protocol (method and experiments sections): no description is given of how the training measurement-support pairs are generated relative to the test distribution (sparsity level, noise statistics, signal class). The generalization assumption required for the fixed CSEN to deliver the claimed accuracy on unseen signals is therefore untested; a domain-shift experiment or cross-validation across signal classes is needed.

    Authors: We will expand the method and experiments sections to explicitly describe how the training measurement-support pairs are generated, confirming that they follow the same sparsity levels, noise statistics, and signal classes as the test distribution drawn from the benchmark datasets. In addition, we will add a domain-shift or cross-validation experiment across signal classes to directly test the generalization assumption. revision: yes

Circularity Check

0 steps flagged

No circularity detected in CSEN derivation or claims

full rationale

The paper presents CSEN as a convolutional neural network trained in a standard supervised manner on pairs of compressive measurements and support sets to learn a direct mapping for support estimation. Performance is evaluated empirically on benchmark datasets rather than derived from any self-referential equations, fitted parameters renamed as predictions, or load-bearing self-citations. No derivation chain reduces the claimed SOTA results or complexity reduction to inputs by construction; the approach is self-contained as an empirical ML method with independent test-set evaluation.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review; no free parameters, axioms, or invented entities are specified in the provided text.

pith-pipeline@v0.9.0 · 5750 in / 970 out tokens · 22168 ms · 2026-05-24T15:00:51.028585+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Advance Warning Methodologies for COVID-19 using Chest X-Ray Images

    eess.IV 2020-06 unverdicted novelty 4.0

    Introduces the Early-QaTa-COV19 dataset and reports that CSEN reaches over 97% sensitivity and over 95.5% specificity for early COVID-19 detection from X-rays.

  2. Convolutional Sparse Support Estimator Based Covid-19 Recognition from X-ray Images

    eess.IV 2020-05 unverdicted novelty 4.0

    Introduces CSEN, a non-iterative network bridging sparse representation and deep learning, for Covid-19 detection from X-ray images with limited training data.

Reference graph

Works this paper leans on

71 extracted references · 71 canonical work pages · cited by 2 Pith papers · 2 internal anchors

  1. [1]

    Compressed sensing,

    D. L. Donoho et al. , “Compressed sensing,” IEEE Transactions on information theory, vol. 52, no. 4, pp. 1289–1306, 2006

    work page 2006

  2. [2]

    Compressive sampling,

    E. J. Cand `es et al. , “Compressive sampling,” in Proceedings of the International Congress of Mathematicians, vol. 3, 2006, pp. 1433–1452

    work page 2006

  3. [3]

    On projection matrix optimization for compressive sensing systems,

    G. Li, Z. Zhu, D. Yang, L. Chang, and H. Bai, “On projection matrix optimization for compressive sensing systems,” IEEE Transactions on Signal Processing, vol. 61, no. 11, pp. 2887–2898, 2013

    work page 2013

  4. [4]

    Elad, Sparse and redundant representations: from theory to appli- cations in signal and image processing

    M. Elad, Sparse and redundant representations: from theory to appli- cations in signal and image processing . Springer Science & Business Media, 2010

    work page 2010

  5. [5]

    Information- theoretic limits on sparse support recovery: Dense versus sparse mea- surements,

    W. Wang, M. J. Wainwright, and K. Ramchandran, “Information- theoretic limits on sparse support recovery: Dense versus sparse mea- surements,” in 2008 IEEE International Symposium on Information Theory. IEEE, 2008, pp. 2197–2201

    work page 2008

  6. [6]

    Robust support recovery using sparse compressive sensing matrices,

    J. Haupt and R. Baraniuk, “Robust support recovery using sparse compressive sensing matrices,” in 2011 45th Annual Conference on Information Sciences and Systems . IEEE, 2011, pp. 1–6

    work page 2011

  7. [7]

    Sampling bounds for sparse support recov- ery in the presence of noise,

    G. Reeves and M. Gastpar, “Sampling bounds for sparse support recov- ery in the presence of noise,” in 2008 IEEE International Symposium on Information Theory . IEEE, 2008, pp. 2187–2191

    work page 2008

  8. [8]

    Signal recovery from random mea- surements via orthogonal matching pursuit,

    J. A. Tropp and A. C. Gilbert, “Signal recovery from random mea- surements via orthogonal matching pursuit,” IEEE Transactions on information theory, vol. 53, no. 12, pp. 4655–4666, 2007

    work page 2007

  9. [9]

    Cosamp: Iterative signal recovery from in- complete and inaccurate samples,

    D. Needell and J. A. Tropp, “Cosamp: Iterative signal recovery from in- complete and inaccurate samples,” Applied and computational harmonic analysis, vol. 26, no. 3, pp. 301–321, 2009

    work page 2009

  10. [10]

    Limits on support recovery with probabilistic models: An information-theoretic framework,

    J. Scarlett and V . Cevher, “Limits on support recovery with probabilistic models: An information-theoretic framework,” IEEE Transactions on Information Theory, vol. 63, no. 1, pp. 593–620, 2016

    work page 2016

  11. [11]

    Sparse representation for computer vision and pattern recognition,

    J. Wright, Y . Ma, J. Mairal, G. Sapiro, T. S. Huang, and S. Yan, “Sparse representation for computer vision and pattern recognition,” Proceedings of the IEEE , vol. 98, no. 6, pp. 1031–1044, 2010

    work page 2010

  12. [12]

    Robust face recognition via sparse representation,

    J. Wright, A. Y . Yang, A. Ganesh, S. S. Sastry, and Y . Ma, “Robust face recognition via sparse representation,” IEEE transactions on pattern analysis and machine intelligence , vol. 31, no. 2, pp. 210–227, 2008

    work page 2008

  13. [13]

    Efficient spectrum availability information recovery for wideband dsa networks: A weighted compressive sampling approach,

    B. Khalfi, B. Hamdaoui, M. Guizani, and N. Zorba, “Efficient spectrum availability information recovery for wideband dsa networks: A weighted compressive sampling approach,” IEEE Transactions on Wireless Com- munications, vol. 17, no. 4, pp. 2162–2172, 2018

    work page 2018

  14. [14]

    Compressed wideband spec- trum sensing: Concept, challenges, and enablers,

    B. Hamdaoui, B. Khalfi, and M. Guizani, “Compressed wideband spec- trum sensing: Concept, challenges, and enablers,”IEEE Communications Magazine, vol. 56, no. 4, pp. 136–141, 2018

    work page 2018

  15. [15]

    A compressive sensing data acquisition and imaging method for stepped frequency GPRs,

    A. C. Gurbuz, J. H. McClellan, and W. R. Scott, “A compressive sensing data acquisition and imaging method for stepped frequency GPRs,”IEEE Transactions on Signal Processing, vol. 57, no. 7, pp. 2640–2650, 2009

    work page 2009

  16. [16]

    Fully convolutional networks for semantic segmentation,

    J. Long, E. Shelhamer, and T. Darrell, “Fully convolutional networks for semantic segmentation,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2015, pp. 3431–3440

    work page 2015

  17. [17]

    Recon- net: Non-iterative reconstruction of images from compressively sensed measurements,

    K. Kulkarni, S. Lohit, P. Turaga, R. Kerviche, and A. Ashok, “Recon- net: Non-iterative reconstruction of images from compressively sensed measurements,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2016, pp. 449–458

    work page 2016

  18. [18]

    Amp-inspired deep net- works for sparse linear inverse problems,

    M. Borgerding, P. Schniter, and S. Rangan, “Amp-inspired deep net- works for sparse linear inverse problems,” IEEE Transactions on Signal Processing, vol. 65, no. 16, pp. 4293–4308, 2017

    work page 2017

  19. [19]

    Message-passing algo- rithms for compressed sensing,

    D. L. Donoho, A. Maleki, and A. Montanari, “Message-passing algo- rithms for compressed sensing,” Proceedings of the National Academy of Sciences, vol. 106, no. 45, pp. 18 914–18 919, 2009

    work page 2009

  20. [20]

    From few to many: Illumination cone models for face recognition under variable lighting and pose,

    A. S. Georghiades, P. N. Belhumeur, and D. J. Kriegman, “From few to many: Illumination cone models for face recognition under variable lighting and pose,” IEEE Transactions on Pattern Analysis & Machine Intelligence, no. 6, pp. 643–660, 2001

    work page 2001

  21. [21]

    Deep learning face attributes in the wild,

    Z. Liu, P. Luo, X. Wang, and X. Tang, “Deep learning face attributes in the wild,” in Proceedings of International Conference on Computer Vision (ICCV), December 2015

    work page 2015

  22. [22]

    Sparse representation or collaborative representation: Which helps face recognition?

    L. Zhang, M. Yang, and X. Feng, “Sparse representation or collaborative representation: Which helps face recognition?” in 2011 International conference on computer vision . IEEE, 2011, pp. 471–478

    work page 2011

  23. [23]

    On the use of a priori information for sparse signal approximations,

    O. D. Escoda, L. Granai, and P. Vandergheynst, “On the use of a priori information for sparse signal approximations,” IEEE transactions on signal processing, vol. 54, no. 9, pp. 3468–3482, 2006

    work page 2006

  24. [24]

    Recursive recovery of sparse signal sequences from compressive measurements: A review,

    N. Vaswani and J. Zhan, “Recursive recovery of sparse signal sequences from compressive measurements: A review,” IEEE Transactions on Signal Processing, vol. 64, no. 13, pp. 3523–3549, 2016

    work page 2016

  25. [25]

    Optimally sparse representation in general (nonorthogonal) dictionaries via ell1 minimization,

    D. L. Donoho and M. Elad, “Optimally sparse representation in general (nonorthogonal) dictionaries via ell1 minimization,” Proceedings of the National Academy of Sciences , vol. 100, no. 5, pp. 2197–2202, 2003

    work page 2003

  26. [27]

    The restricted isometry property and its implications for compressed sensing,

    E. J. Candes, “The restricted isometry property and its implications for compressed sensing,” Comptes rendus mathematique, vol. 346, no. 9-10, pp. 589–592, 2008

    work page 2008

  27. [28]

    Atomic decomposition by basis pursuit,

    S. S. Chen, D. L. Donoho, and M. A. Saunders, “Atomic decomposition by basis pursuit,” SIAM review, vol. 43, no. 1, pp. 129–159, 2001

    work page 2001

  28. [29]

    The dantzig selector: Statistical estimation when p is much larger than n,

    E. Candes, T. Tao et al. , “The dantzig selector: Statistical estimation when p is much larger than n,” The annals of Statistics , vol. 35, no. 6, pp. 2313–2351, 2007

    work page 2007

  29. [30]

    Stable signal recovery from incomplete and inaccurate measurements,

    E. J. Candes, J. K. Romberg, and T. Tao, “Stable signal recovery from incomplete and inaccurate measurements,” Communications on Pure and Applied Mathematics: A Journal Issued by the Courant Institute of Mathematical Sciences , vol. 59, no. 8, pp. 1207–1223, 2006

    work page 2006

  30. [31]

    Y . C. Eldar and G. Kutyniok, Compressed sensing: theory and applica- tions. Cambridge University Press, 2012

    work page 2012

  31. [32]

    On the lasso and dantzig selector equiv- alence,

    M. S. Asif and J. Romberg, “On the lasso and dantzig selector equiv- alence,” in 2010 44th Annual Conference on Information Sciences and Systems (CISS). IEEE, 2010, pp. 1–6

    work page 2010

  32. [33]

    Regression shrinkage and selection via the lasso,

    R. Tibshirani, “Regression shrinkage and selection via the lasso,” Jour- nal of the Royal Statistical Society: Series B (Methodological) , vol. 58, no. 1, pp. 267–288, 1996

    work page 1996

  33. [34]

    A probabilistic and ripless theory of compressed sensing,

    E. J. Candes and Y . Plan, “A probabilistic and ripless theory of compressed sensing,” IEEE transactions on information theory , vol. 57, no. 11, pp. 7235–7254, 2011

    work page 2011

  34. [35]

    Compressed anomaly detection with multiple mixed observations,

    N. Durgin, R. Grotheer, C. Huang, S. Li, A. Ma, D. Needell, and J. Qin, “Compressed anomaly detection with multiple mixed observations,” in Research in Data Science . Springer, 2019, pp. 211–237

    work page 2019

  35. [36]

    Malicious users discrimination in organized attacks using structured sparsity,

    M. Yamac ¸, B. Sankur, and A. T. Cemgil, “Malicious users discrimination in organized attacks using structured sparsity,” in 2017 25th European Signal Processing Conference (EUSIPCO) . IEEE, 2017, pp. 266–270

    work page 2017

  36. [37]

    Multiuser detection via compressive sensing,

    B. Shim and B. Song, “Multiuser detection via compressive sensing,” IEEE Communications Letters , vol. 16, no. 7, pp. 972–974, 2012

    work page 2012

  37. [38]

    Multiuser detector for uplink grant free noma systems based on modified subspace pursuit algorithm,

    O. O. Oyerinde, “Multiuser detector for uplink grant free noma systems based on modified subspace pursuit algorithm,” in 2018 12th Interna- tional Conference on Signal Processing and Communication Systems (ICSPCS). IEEE, 2018, pp. 1–6

    work page 2018

  38. [39]

    Dynamic compressive sensing-based multi-user detection for uplink grant-free noma,

    B. Wang, L. Dai, Y . Zhang, T. Mir, and J. Li, “Dynamic compressive sensing-based multi-user detection for uplink grant-free noma,” IEEE Communications Letters, vol. 20, no. 11, pp. 2320–2323, 2016

    work page 2016

  39. [40]

    Through the wall target detection/monitoring from compressively sensed signals via structural sparsity,

    M. Yamac ¸, M. Orhan, B. Sankur, A. S. Turk, and M. Gabbouj, “Through the wall target detection/monitoring from compressively sensed signals via structural sparsity,” in 5th International Workshop on Compressed Sensing applied to Radar, Multimodal Sensing,and Imaging , 2018. 12

    work page 2018

  40. [41]

    Nearly sharp sufficient conditions on exact sparsity pattern recovery,

    K. R. Rad, “Nearly sharp sufficient conditions on exact sparsity pattern recovery,” IEEE Transactions on Information Theory , vol. 57, no. 7, pp. 4672–4679, 2011

    work page 2011

  41. [42]

    Information-theoretic bounds on sparsity recovery in the high-dimensional and noisy setting,

    M. Wainwright, “Information-theoretic bounds on sparsity recovery in the high-dimensional and noisy setting,” in 2007 IEEE International Symposium on Information Theory . IEEE, 2007, pp. 961–965

    work page 2007

  42. [43]

    Approximate sparsity pattern recovery: Information-theoretic lower bounds,

    G. Reeves and M. C. Gastpar, “Approximate sparsity pattern recovery: Information-theoretic lower bounds,” IEEE Transactions on Information Theory, vol. 59, no. 6, pp. 3451–3465, 2013

    work page 2013

  43. [44]

    Near-ideal model selection by ell1 minimization,

    E. J. Cand `es, Y . Plan et al. , “Near-ideal model selection by ell1 minimization,” The Annals of Statistics, vol. 37, no. 5A, pp. 2145–2177, 2009

    work page 2009

  44. [45]

    Orthogonal matching pursuit for sparse signal recovery with noise

    T. T. Cai and L. Wang, “Orthogonal matching pursuit for sparse signal recovery with noise.” Institute of Electrical and Electronics Engineers, 2011

    work page 2011

  45. [46]

    The sampling rate-distortion tradeoff for sparsity pattern recovery in compressed sensing,

    G. Reeves and M. Gastpar, “The sampling rate-distortion tradeoff for sparsity pattern recovery in compressed sensing,” IEEE Transactions on Information Theory, vol. 58, no. 5, pp. 3065–3092, 2012

    work page 2012

  46. [47]

    Necessary and sufficient conditions for sparsity pattern recovery,

    A. K. Fletcher, S. Rangan, and V . K. Goyal, “Necessary and sufficient conditions for sparsity pattern recovery,” IEEE Transactions on Infor- mation Theory, vol. 55, no. 12, pp. 5758–5772, 2009

    work page 2009

  47. [48]

    Learning sparse representations for human action recognition,

    T. Guha and R. K. Ward, “Learning sparse representations for human action recognition,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 34, no. 8, pp. 1576–1588, 2011

    work page 2011

  48. [49]

    A survey on representation-based classification and detection in hyperspectral remote sensing imagery,

    W. Li and Q. Du, “A survey on representation-based classification and detection in hyperspectral remote sensing imagery,” Pattern Recognition Letters, vol. 83, pp. 115–123, 2016

    work page 2016

  49. [50]

    Modified-cs: Modifying compressive sensing for problems with partially known support,

    N. Vaswani and W. Lu, “Modified-cs: Modifying compressive sensing for problems with partially known support,” IEEE Transactions on Signal Processing, vol. 58, no. 9, pp. 4595–4607, 2010

    work page 2010

  50. [51]

    Sparsity and incoherence in compressive sampling,

    E. Candes and J. Romberg, “Sparsity and incoherence in compressive sampling,” Inverse problems, vol. 23, no. 3, p. 969, 2007

    work page 2007

  51. [52]

    Observed universality of phase transitions in high-dimensional geometry, with implications for modern data analysis and signal processing,

    D. Donoho and J. Tanner, “Observed universality of phase transitions in high-dimensional geometry, with implications for modern data analysis and signal processing,” Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, vol. 367, no. 1906, pp. 4273–4293, 2009

    work page 1906

  52. [53]

    Learning fast approximations of sparse coding,

    K. Gregor and Y . LeCun, “Learning fast approximations of sparse coding,” in Proceedings of the 27th International Conference on In- ternational Conference on Machine Learning . Omnipress, 2010, pp. 399–406

    work page 2010

  53. [54]

    Fast ell1-minimization algorithms for robust face recognition,

    A. Y . Yang, Z. Zhou, A. G. Balasubramanian, S. S. Sastry, and Y . Ma, “Fast ell1-minimization algorithms for robust face recognition,” IEEE Transactions on Image Processing, vol. 22, no. 8, pp. 3234–3246, 2013

    work page 2013

  54. [55]

    TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems

    M. Abadi, A. Agarwal, P. Barham, E. Brevdo, Z. Chen, C. Citro, G. S. Corrado, A. Davis, J. Dean, M. Devin et al. , “Tensorflow: Large-scale machine learning on heterogeneous distributed systems,” arXiv preprint arXiv:1603.04467, 2016

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  55. [56]

    Adam: A Method for Stochastic Optimization

    D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” arXiv preprint arXiv:1412.6980 , 2014

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  56. [57]

    Dlib-ml: A machine learning toolkit,

    D. E. King, “Dlib-ml: A machine learning toolkit,” Journal of Machine Learning Research, vol. 10, no. Jul, pp. 1755–1758, 2009

    work page 2009

  57. [58]

    Distributed optimization and statistical learning via the alternating direction method of multipliers,

    S. Boyd, N. Parikh, E. Chu, B. Peleato, J. Eckstein et al., “Distributed optimization and statistical learning via the alternating direction method of multipliers,” Foundations and Trends R⃝ in Machine learning , vol. 3, no. 1, pp. 1–122, 2011

    work page 2011

  58. [59]

    Homotopy continuation for sparse signal representation,

    D. M. Malioutov, M. Cetin, and A. S. Willsky, “Homotopy continuation for sparse signal representation,” in Proceedings.(ICASSP’05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005., vol. 5. IEEE, 2005, pp. v–733

    work page 2005

  59. [60]

    Gradient projection for sparse reconstruction: Application to compressed sensing and other inverse problems,

    M. A. Figueiredo, R. D. Nowak, and S. J. Wright, “Gradient projection for sparse reconstruction: Application to compressed sensing and other inverse problems,” IEEE Journal of selected topics in signal processing , vol. 1, no. 4, pp. 586–597, 2007

    work page 2007

  60. [61]

    An interior-point method for large- scale l1-regularized logistic regression,

    K. Koh, S.-J. Kim, and S. Boyd, “An interior-point method for large- scale l1-regularized logistic regression,” Journal of Machine learning research, vol. 8, no. Jul, pp. 1519–1555, 2007

    work page 2007

  61. [62]

    l1-magic: Recovery of sparse signals via convex programming,

    E. Candes and J. Romberg, “l1-magic: Recovery of sparse signals via convex programming,” URL: www. acm. caltech. edu/l1magic/downloads/l1magic. pdf, vol. 4, p. 14, 2005

    work page 2005

  62. [63]

    Solving inverse problems with piecewise linear estimators: From gaussian mixture models to structured sparsity,

    G. Yu, G. Sapiro, and S. Mallat, “Solving inverse problems with piecewise linear estimators: From gaussian mixture models to structured sparsity,” IEEE Transactions on Image Processing , vol. 21, no. 5, pp. 2481–2499, 2012

    work page 2012

  63. [64]

    Model- based compressive sensing,

    R. G. Baraniuk, V . Cevher, M. F. Duarte, and C. Hegde, “Model- based compressive sensing,” IEEE Transactions on Information Theory , vol. 56, no. 4, pp. 1982–2001, 2010

    work page 1982

  64. [65]

    Microlocal analysis of the geometric sep- aration problem,

    D. Donoho and G. Kutyniok, “Microlocal analysis of the geometric sep- aration problem,” Communications on Pure and Applied Mathematics , vol. 66, no. 1, pp. 1–47, 2013

    work page 2013

  65. [66]

    Structured sparsity: from mixed norms to structured shrinkage,

    M. Kowalski and B. Torr ´esani, “Structured sparsity: from mixed norms to structured shrinkage,” in SPARS’09-Signal Processing with Adaptive Sparse Structured Representations, 2009

    work page 2009

  66. [67]

    Matching pursuits with time-frequency dictionaries,

    S. G. Mallat and Z. Zhang, “Matching pursuits with time-frequency dictionaries,” IEEE Transactions on signal processing , vol. 41, no. 12, pp. 3397–3415, 1993

    work page 1993

  67. [68]

    Bayesian compressive sensing,

    S. Ji, Y . Xue, L. Carin et al. , “Bayesian compressive sensing,” IEEE Transactions on signal processing , vol. 56, no. 6, p. 2346, 2008

    work page 2008

  68. [69]

    Nonlin- ear wavelet image processing: variational problems, compression, and noise removal through wavelet shrinkage,

    A. Chambolle, R. A. De V ore, N.-Y . Lee, and B. J. Lucier, “Nonlin- ear wavelet image processing: variational problems, compression, and noise removal through wavelet shrinkage,” IEEE Transactions on Image Processing, vol. 7, no. 3, pp. 319–335, 1998

    work page 1998

  69. [70]

    Com- pressively sensed image recognition,

    A. De ˘gerli, S. Aslan, M. Yamac, B. Sankur, and M. Gabbouj, “Com- pressively sensed image recognition,” in 2018 7th European Workshop on Visual Information Processing (EUVIP) . IEEE, 2018, pp. 1–6

    work page 2018

  70. [71]

    Direct inference on compressive measurements using convolutional neural networks,

    S. Lohit, K. Kulkarni, and P. Turaga, “Direct inference on compressive measurements using convolutional neural networks,” in 2016 IEEE International Conference on Image Processing (ICIP) , Sep. 2016, pp. 1913–1917

    work page 2016

  71. [72]

    Learning to invert: Signal recovery via deep convolutional networks,

    A. Mousavi and R. G. Baraniuk, “Learning to invert: Signal recovery via deep convolutional networks,” in 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2017, pp. 2272–2276

    work page 2017