Pith. sign in

REVIEW 2 major objections 4 minor 1 cited by

A Nonconvex Approach for Exact and Efficient Multichannel Sparse Blind Deconvolution

T0 review · 2 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read This paper proves that plain Riemannian gradient descent, run from a random initialization on a preconditioned Huber-loss sphere, exactly recovers both the convolution kernel and the sparse inputs in multichannel sparse blind…

desk verdict Substantial and likely correct theory for multichannel sparse blind deconvolution, but the advertised Phase-2 iteration count omits a theta-dependence that its own sharpness analysis forces. read the letter →

arxiv 1908.10776 v3 pith:ZE4R75XM submitted 2019-08-28 eess.SP cs.LGeess.IVmath.OCstat.ML

Qing Qu , Xiao Li , Zhihui Zhu This is my paper · ORCID
classification eess.SPcs.LGeess.IVmath.OCstat.ML
keywords nonconvexoptimizationblinddeconvolutionsparsityRiemanniangradientdescentBernoulli-GaussianmodelHuberlosspreconditioningsamplecomplexity
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper sets out to prove that a straightforward algorithm—plain (vanilla) Riemannian gradient descent from a random point on the unit sphere, run on a smoothed $\ell^1$ (Huber) loss with data preconditioning—recovers both the unknown kernel and the sparse input signals in multichannel sparse blind deconvolution, up to the unavoidable signed-shift ambiguity. Previous nonconvex guarantees needed noisy gradient steps or tolerated far worse sample and time complexity; this paper claims the first provable guarantee for vanilla first-order methods with random initialization. If the result holds, practitioners can use a simple gradient loop with no tunable noise schedule and still get exact recovery, provided the inputs are independent Bernoulli-Gaussian with known sparsity and the kernel has no zero Fourier entries. The proof works by showing that, after preconditioning, the objective's geometry is benign: every region of the sphere around a target solution has gradients pointing toward that solution, and a second LP-rounding phase sharpens the approximate solution to exact precision.

What carries the argument

The argument rests on three objects that work together. The preconditioner $P = (\frac{1}{\theta n p}\sum_{i=1}^p C_{y_i}^{\mathsf T} C_{y_i})^{-1/2}$ approximately orthogonalizes the circulant convolution operator: it makes $R = C_a P$ close to the orthogonal matrix $Q = C_a (C_a^{\mathsf T} C_a)^{-1/2}$, symmetrizing the landscape and removing the ill-conditioning caused by a generic kernel. The Huber loss $h_\mu(z)$, a first-order smooth surrogate of the $\ell^1$ norm, promotes sparsity while keeping gradients Lipschitz, so random-process concentration and perturbation arguments go through. Finally, the $2n$ spherical regions $S_i^\xi = \{ q \in S^{n-1} : |q_i|/\|q_{-i}\|_\infty \ge \sqrt{1+\xi},\ q_i \neq 0\}$, each containing one target solution, allow the analysis to establish a regularity condition (the negative gradient aligns with the direction to $e_i$), an implicit-regularization bound (iterates never leave their region), and then a sharpness bound for the LP-rounding objective that gives linear convergence of the second phase.

What would settle it

Generate measurements $y_i = a * x_i$ with $a$ an invertible kernel of condition number $\kappa$, $x_i$ i.i.d. Bernoulli-Gaussian at $\theta=0.25$, and choose $p$ above the Theorem 3.1 bound for small $n$ (say $n=64$, $\theta=0.25$, $\mu=10^{-2}$); the two-stage algorithm should recover the kernel and signals to $\varepsilon$-precision with probability at least $1/2$. If it systematically fails at such parameters—for example, by converging to a spurious stationary point—the claimed guarantee is false; the same test with a kernel whose DFT contains a zero probes the boundary of the invertibility assumption rather than the core claim.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that multichannel sparse blind deconvolution has benign global geometry for a smooth sparsity surrogate, not for the heavy-tailed $\ell^4$ loss used in earlier nonconvex work. For an invertible kernel $a$ with condition number $\kappa$ and i.i.d. Bernoulli-Gaussian inputs with sparsity $\theta \in (1/n, 1/3)$, the preconditioned Huber objective over the sphere satisfies a regularity condition in each region $S_i^\xi$: the Riemannian gradient has inner product at least proportional to $\|q - e_i\|$ with the direction to the target basis vector $e_i$, and an implicit-regularization bound keeps iterates inside the region. With probability at least $1/2$ over a uniform random initialization, vanilla Riemannian gradient descent therefore converges linearly to an $O(\mu)$-accurate solution in $O(\theta^{-1} n^4 \log(1/\mu) + \log(np)\log(1/\varepsilon))$ iterations, and a projected subgradient method on an LP-rounding problem converges linearly to the exact kernel and signals up to a signed shift. The sample complexity is $p \ge C\max\{n, \kappa^8/(\theta \mu^2 \sigma_{\min}^2) \log^4 n\}\, \theta^{-2} n^4 \log^3 n \log(\theta n/\mu)$, an improvement over the prior $p \ge \tilde{\Omega}(\max\{n,\kappa^8\} n^8/\varepsilon^8)$.

Load-bearing premise

The load-bearing premise is the generative model: the $p$ inputs are independent Bernoulli-Gaussian with known sparsity $\theta$ in $(1/n, 1/3)$, and the kernel's discrete Fourier transform has no zero entries; if the signals are not independent sparse Gaussians or the kernel has deep frequency nulls, the benign geometry and the recovery guarantee can fail.

Editorial extensions

If this is right

  • Vanilla Riemannian gradient descent from a random initialization converges linearly to an approximate solution, so the first phase has no noise-injection or step-size-schedule parameters to tune.
  • The two-stage method achieves exact recovery: the phase-two LP rounding converges linearly to the signed-shift target, removing the $O(\mu)$ smoothing error introduced by the Huber loss.
  • The sample complexity is polynomial in $n$ (about $n^4$) and independent of the recovery precision $\varepsilon$, improving on the previous $\tilde{\Omega}(n^8/\varepsilon^8)$ bound; the iteration count is about $n^4 + \log(1/\varepsilon)$.
  • With FFT implementation, each gradient or subgradient update costs $O(p n \log n)$, so the method is computationally practical for large $n$ and is demonstrated on 2D STORM super-resolution imaging data.
  • Constant sparsity levels $\theta$ up to about $1/3$ are covered, in contrast to earlier convex approaches requiring $\theta = O(1/\sqrt{n})$.

Reading between the lines

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

  • The paper's phase-transition plots show successful recovery with $p$ close to poly-logarithmic in $n$, far below the theorem's $n^4$ bound; my inference is that the sample-complexity statement is loose and a sharper concentration argument could reduce the polynomial gap.
  • The region-based geometry plus preconditioning is likely transferable to convolutional dictionary learning and blind gain-and-phase calibration, where similar bilinear nonconvex landscapes appear; the paper suggests these directions but does not prove them.
  • If the Bernoulli-Gaussian assumption is replaced by heavier-tailed or dependent sparse models, the concentration and sharpness arguments would need reworking; a natural testable extension is to determine empirically how far the phase transition degrades under correlated supports.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 4 minor

Summary. The paper studies multichannel sparse blind deconvolution (MCS-BD), where one observes p circulant convolutions y_i = a * x_i with an invertible kernel a and i.i.d. Bernoulli-Gaussian sparse inputs x_i, and aims to recover a and all x_i up to the intrinsic signed-shift ambiguity. The authors formulate a smooth Huber-loss objective on the sphere with a data-dependent preconditioner and prove that a vanilla Riemannian gradient descent (RGD) method from a random initialization reaches an O(mu) neighborhood of the target, after which a projected subgradient method for an LP rounding problem converges linearly to the exact solution. The main theorem gives sample-complexity and iteration-complexity bounds, and the paper reports experiments on synthetic data and on a STORM microscopy dataset. The proof is developed in Appendices C-H through population geometry, concentration inequalities, and a perturbation analysis around RQ^{-1} approximately equal to I.

Significance. If the result is correct in its stated quantitative form, this is a significant contribution: it is the first provable guarantee that vanilla first-order methods with random initialization solve MCS-BD exactly (up to the signed-shift ambiguity), and its sample complexity improves on the prior nonconvex result of Li and Bresler. The paper is unusually careful in structure: the geometric assumptions are explicit, the regularity and implicit-regularization conditions are stated as separate propositions, the appendices contain a substantial amount of proof detail, and the authors provide code and honest discussion of the looseness of their bounds. The central recovery claim is not affected by the issue I raise below; however, the headline efficiency statement in Theorem 3.1 overstates what the Phase-2 sharpness analysis actually proves, and this needs to be corrected before the paper can be accepted in its present form.

major comments (2)
  1. [Theorem 3.1 / Appendix D.3 (Proposition D.9)] The Phase-2 iteration count claimed in Theorem 3.1, namely C log(np) log(1/epsilon), does not follow from the sharpness analysis in Appendix D.3. Proposition D.9 sets the sharpness constant to alpha = (1/50) sqrt(2/pi) theta and the projected subgradient norm bound to beta = 36 log(np), and then chooses eta with eta^2 >= 1 - alpha^2/(2 beta^2). Consequently 1 - eta is proportional to theta^2 / log^2(np), so the number of projected subgradient iterations needed to reduce dist(q) from a constant to epsilon is k >= C theta^{-2} log^2(np) log(1/epsilon), not C log(np) log(1/epsilon). Proposition 3.5 conceals this dependence inside the unspecified constant c2 in the lower bound on eta. This is a load-bearing quantitative claim: the theorem as stated overstates the Phase-2 efficiency by a factor of theta^{-2} log(np). The recovery conclusion is not invalidated, but the complexity statement and Proposition 3.5 must be revised to state the theta and log factors explicitly.
  2. [Proposition 3.4 / Appendix D.1 (Theorem D.1)] The Phase-1 iteration bound is stated inconsistently across the paper. Theorem 3.1 uses theta^{-1} n^4 log(1/mu), while Proposition 3.4 states N := C theta n^4 log(1/mu) and Theorem D.1 states K := C theta max{n^4, n^{5/2}/mu} log(1/mu). The proof of Theorem D.1 derives contraction factors 1 - tau alpha_1 with tau alpha_1 on the order of theta/n^4 and 1 - tau alpha_2 with tau alpha_2 on the order of mu theta/n, which support the inverse theta dependence of the main theorem rather than the theta factor displayed in Proposition 3.4 and Theorem D.1. Please reconcile these statements so that the main theorem and the appendices report the same bound.
minor comments (4)
  1. [Abstract and keywords] The keyword 'Riemmanian' should be spelled 'Riemannian'.
  2. [Table 1] The time-complexity entry for 'Ours' should be updated after the Phase-2 complexity claim is corrected; as typeset it omits the theta and log(np) factors that the analysis in Appendix D.3 actually yields.
  3. [Appendix G, proof of Proposition G.1] The shorthand notation PqK is used without being defined in Appendix A; Appendix A defines the projection notation P_v and P_{v^\perp}, so the shorthand should either be defined explicitly or replaced by the longer notation.
  4. [Section 4.1] The experiments use Riemannian line search rather than the fixed step sizes prescribed by Theorem 3.1; this is reasonable for illustration, but the text should state explicitly that the empirical convergence plots are not a direct test of the theorem's step-size and iteration-count prescriptions.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the recovery guarantee is derived from an explicit statistical model, with self-citations serving only as ancillary supporting lemmas.

full rationale

The paper's central claim is a probabilistic recovery guarantee for a vanilla first-order method, not a fitted or self-referential construction. The statistical assumptions (invertible kernel a and i.i.d. Bernoulli-Gaussian inputs with sparsity parameter theta) are stated in Section 2.1 and are then used to compute population gradients, concentration bounds, and sharpness constants in Appendices E-G; the conclusion is not used to define these quantities. The preconditioner P in (6) is data-dependent, but it is derived from the observed covariance and analyzed through independent perturbation bounds in Appendix H; it is not selected to force the target solution. The two-stage convergence proof is self-contained: Phase 1 uses the regularity condition (11) and implicit regularization (12), both proved from the BG model, and Phase 2 uses an explicitly computed sharpness constant alpha = (1/50) sqrt(2/pi) theta, not a fitted parameter. The self-citations, mainly [SQW16] and [SQW17], appear as supporting lemmas such as Lemma B.5 for lower-bounding sparse random matrices; those lemmas do not assume the MCS-BD recovery result and are not invoked to prohibit alternative approaches. The skeptical concern about the Phase-2 iteration count possibly omitting a theta^{-2} log^2(np) factor is an internal consistency issue in the stated complexity bound, not a circular reduction of the claimed result to its own inputs. Accordingly, no circular step is established under the required standard.

Assumptions & free parameters 1 free parameters · 6 assumptions · 0 invented entities

The proof rests on standard concentration inequalities, an external sparse-subspace lemma from [SQW17], and the statistical model assumptions. No invented physical entities are introduced; the only new objects are mathematical (Huber-loss formulation, preconditioned variables), which do not qualify as invented entities. The single algorithm hyperparameter is the Huber smoothing mu.

free parameters (1)
  • mu (Huber smoothing parameter) = mu = 10^{-2} in experiments; theory requires mu <= c min{theta, 1/sqrt(n)}
    Controls the tradeoff between smoothness and Phase-1 accuracy; the final exact recovery via LP rounding requires mu <= 1/25. It is a hyperparameter of the algorithm, not fitted to the data, but the theoretical guarantee depends on it.
assumptions (6)
  • domain assumption The kernel a is invertible: its DFT has no zero entries, with condition number kappa = max|a_hat_i| / min|a_hat_i|.
    Section 2.1, Assumption 1. Needed for C_a invertible, to define inverse kernel h and the preconditioning analysis (Lemmas H.2-H.4).
  • domain assumption Inputs x_i are i.i.d. Bernoulli-Gaussian with sparsity theta in (1/n, 1/3).
    Section 2.1, Assumption 2. Used throughout the population geometry (Appendix E) and concentration proofs; the theorem does not cover other signal distributions.
  • domain assumption The preconditioning matrix P uses the sparsity parameter theta (eq 6); the proof requires this, though footnote 9 notes scaling is immaterial.
    Eq (6) and footnote 9. If theta is unknown, P is only defined up to scale, which does not change the landscape, but the theoretical statement assumes the correct normalization for the concentration of R to Q.
  • standard math Random initialization is uniform on S^{n-1}; Lemma D.3 (from BJS18, GBW18) ensures it lands in one of the 2n regions S_i^+- with probability at least 1/2.
    Section 3.2, Lemma D.3. This external geometric measure result is cited, not proved in this paper.
  • standard math Sparse-subspace lemma B.5 from [SQW17] (control of ell-1 norms of sub-Gaussian matrices) is used to prove sharpness of the LP rounding objective.
    Lemma B.5, used in Lemma D.8. This external result is load-bearing for the linear convergence of Phase 2.
  • standard math Standard concentration inequalities (Bernstein, Gaussian concentration, moment-control) and matrix perturbation bounds are applied without proof.
    Appendix B lists these as tools; they are standard background for the proof.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Nonconvex Approach for Exact and Efficient Multichannel Sparse Blind Deconvolution." pith.science (2026). https://pith.science/paper/ZE4R75XM

@misc{pith2026190810776,
  author       = {Pith},
  title        = {Pith review of: A Nonconvex Approach for Exact and Efficient Multichannel Sparse Blind Deconvolution},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZE4R75XM}},
  note         = {Machine review of arXiv:1908.10776}
}
abstract

We study the multi-channel sparse blind deconvolution (MCS-BD) problem, whose task is to simultaneously recover a kernel $\mathbf a$ and multiple sparse inputs $\{\mathbf x_i\}_{i=1}^p$ from their circulant convolution $\mathbf y_i = \mathbf a \circledast \mathbf x_i $ ($i=1,\cdots,p$). We formulate the task as a nonconvex optimization problem over the sphere. Under mild statistical assumptions of the data, we prove that the vanilla Riemannian gradient descent (RGD) method, with random initializations, provably recovers both the kernel $\mathbf a$ and the signals $\{\mathbf x_i\}_{i=1}^p$ up to a signed shift ambiguity. In comparison with state-of-the-art results, our work shows significant improvements in terms of sample complexity and computational efficiency. Our theoretical results are corroborated by numerical experiments, which demonstrate superior performance of the proposed approach over the previous methods on both synthetic and real datasets.

Figures

Figures reproduced from arXiv: 1908.10776 by the authors.

Figure 1
Figure 1. Comparison of optimization landscapes for differe [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Illustration of the set S 1` ξ in 3-dimension. Here we project the 3D unit sphere onto the plan spanned by e2 and e3. Region 1 (purple region) denotes the interior of S 1` ξ when ξ “ 0, where it includes one unique target solution. In this case, Ť3 i“1 S ˘i ξ forms a full partition of the sphere, and the saddle points (denoted by ˆ) and local maximizers (denoted by ) are on the boundary of the set. When ξ ą 0, the … view at source ↗
Figure 3
Figure 3. Comparison of iterate conver￾gence. p “ 50, n “ 200, θ “ 0.25 [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figures from the paper (2 more)
Figure 5
Figure 5. Figure 5: Comparison of phase transition on pp, nq with fixed θ “ 0.25. Here white denotes successful recovery while black indicates failure. If q‹ achieves the target solution, it should satisfy P q‹ “ sℓ rhs, with sℓ rhs being some circulant shift of the inverse kernel of a an…
Figure 6
Figure 6. Figure 6: STORM imaging via solving MCS-BD. The first line shows (a) observed image, (b) ground truth, (c) recovered image by optimizing Huber-loss, and (d) by ℓ 4 -loss. The second line, (e) ground truth kernel, (f) recovered by optimizing Huber-loss, and (g) by ℓ 4 -loss. 4.2 …

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Short-and-Sparse Deconvolution -- A Geometric Approach

    eess.SP 2019-08 conditional novelty 5.0 of 10

    A practical alternating descent algorithm with data-driven initialization, momentum, homotopy continuation, and reweighting solves short-and-sparse blind deconvolution on synthetic and real imaging and neuroscience da...

Reference graph

Works this paper leans on

89 extracted references · 59 canonical work pages · cited by 1 Pith paper

  1. [1]

    Leveraging diversity and sparsity in blind deconvolution

    Ali Ahmed and Laurent Demanet. Leveraging diversity and sparsity in blind deconvolution. IEEE Transactions on Information Theory , 64(6):3975--4000, 2018

    2018

  2. [2]

    Multichannel blind deconvolution and equalization using the natural gradient

    Shun-ichi Amari, Scott C Douglas, Andrzej Cichocki, and Howard H Yang. Multichannel blind deconvolution and equalization using the natural gradient. In First IEEE Signal Processing Workshop on Signal Processing Advances in Wireless Communications , pages 101--104. IEEE, 1997

    1997

  3. [3]

    Optimization algorithms on matrix manifolds

    P-A Absil, Robert Mahony, and Rodolphe Sepulchre. Optimization algorithms on matrix manifolds . Princeton University Press, 2009

    2009

  4. [4]

    Blind deconvolution using convex programming

    Ali Ahmed, Benjamin Recht, and Justin Romberg. Blind deconvolution using convex programming. IEEE Transactions on Information Theory , 60(3):1711--1732, 2014

    2014

  5. [5]

    Fast convolutional sparse coding

    Hilton Bristow, Anders Eriksson, and Simon Lucey. Fast convolutional sparse coding. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages 391--398, 2013

    2013

  6. [6]

    Weak sharp minima in mathematical programming

    James V Burke and Michael C Ferris. Weak sharp minima in mathematical programming. SIAM Journal on Control and Optimization , 31(5):1340--1359, 1993

    1993

  7. [7]

    Matrix analysis , volume 169

    Rajendra Bhatia. Matrix analysis , volume 169. Springer Science & Business Media, 2013

    2013

  8. [8]

    Subgradient Descent Learns Orthogonal Dictionaries

    Yu Bai, Qijia Jiang, and Ju Sun. Subgradient descent learns orthogonal dictionaries. arXiv preprint arXiv:1810.10702 , 2018

    work page Pith review arXiv 2018

Show all 89 references
  1. [9]

    Imaging intracellular fluorescent proteins at nanometer resolution

    Eric Betzig, George H Patterson, Rachid Sougrat, O Wolf Lindwasser, Scott Olenych, Juan S Bonifacino, Michael W Davidson, Jennifer Lippincott-Schwartz, and Harald F Hess. Imaging intracellular fluorescent proteins at nanometer resolution. Science , 313(5793):1642--1645, 2006

    2006

  2. [10]

    Subgradient methods

    Stephen Boyd, Lin Xiao, and Almir Mutapcic. Subgradient methods. lecture notes of EE392o, Stanford University, Autumn Quarter , 2004:2004--2005, 2003

    2004

  3. [11]

    Harnessing structures in big data via guaranteed low-rank matrix estimation: Recent theory and fast algorithms via convex and nonconvex optimization

    Yudong Chen and Yuejie Chi. Harnessing structures in big data via guaranteed low-rank matrix estimation: Recent theory and fast algorithms via convex and nonconvex optimization. IEEE Signal Processing Magazine , 35(4), 2018

    2018

  4. [12]

    Low-rank matrix recovery with composite optimization: good conditioning and rapid convergence

    Vasileios Charisopoulos, Yudong Chen, Damek Davis, Mateo D \' az, Lijun Ding, and Dmitriy Drusvyatskiy. Low-rank matrix recovery with composite optimization: good conditioning and rapid convergence. arXiv preprint arXiv:1904.10020 , 2019

    1904 arXiv

  5. [13]

    Gradient descent with random initialization: fast global convergence for nonconvex phase retrieval

    Yuxin Chen, Yuejie Chi, Jianqing Fan, and Cong Ma. Gradient descent with random initialization: fast global convergence for nonconvex phase retrieval. Mathematical Programming , 176(1-2):5--37, 2019

    2019

  6. [14]

    Convolutional dictionary learning: Acceleration and convergence

    Il Yong Chun and Jeffrey A Fessler. Convolutional dictionary learning: Acceleration and convergence. IEEE Transactions on Image Processing , 27(4):1697--1712, 2017

    2017

  7. [15]

    Guaranteed blind sparse spikes deconvolution via lifting and convex optimization

    Yuejie Chi. Guaranteed blind sparse spikes deconvolution via lifting and convex optimization. IEEE Journal of Selected Topics in Signal Processing , 10(4):782--794, 2016

    2016

  8. [16]

    Nonconvex optimization meets low-rank matrix factorization: An overview

    Yuejie Chi, Yue M Lu, and Yuxin Chen. Nonconvex optimization meets low-rank matrix factorization: An overview. arXiv preprint arXiv:1809.09573 , 2018

    2018 arXiv

  9. [17]

    Cand \`e s, Xiaodong Li, and Mahdi Soltanolkotabi

    Emmanuel J. Cand \`e s, Xiaodong Li, and Mahdi Soltanolkotabi. Phase retrieval via wirtinger flow: Theory and algorithms. Information Theory, IEEE Transactions on , 61(4):1985--2007, April 2015

    1985

  10. [18]

    A note on the blind deconvolution of multiple sparse signals from unknown subspaces

    Augustin Cosse. A note on the blind deconvolution of multiple sparse signals from unknown subspaces. In Wavelets and Sparsity XVII , volume 10394, page 103941N. International Society for Optics and Photonics, 2017

    2017

  11. [19]

    The null space problem i

    Thomas F Coleman and Alex Pothen. The null space problem i. complexity. SIAM Journal on Algebraic Discrete Methods , 7(4):527--537, 1986

    1986

  12. [20]

    Tail bounds via generic chaining

    Sjoerd Dirksen et al. Tail bounds via generic chaining. Electronic Journal of Probability , 20, 2015

    2015

  13. [21]

    Subgradient methods for sharp weakly convex functions

    Damek Davis, Dmitriy Drusvyatskiy, Kellie J MacPhee, and Courtney Paquette. Subgradient methods for sharp weakly convex functions. Journal of Optimization Theory and Applications , 179(3):962--982, 2018

    2018

  14. [22]

    Decoupling: from dependence to independence

    Victor De la Pena and Evarist Gin \'e . Decoupling: from dependence to independence . Springer Science & Business Media, 2012

    2012

  15. [23]

    Noisy dual principal component pursuit

    Tianyu Ding, Zhihui Zhu, Tianjiao Ding, Yunchen Yang, Daniel Robinson, Manolis Tsakiris, and Rene Vidal. Noisy dual principal component pursuit. In International Conference on Machine Learning , pages 1617--1625, 2019

    2019

  16. [24]

    A blind sparse deconvolution method for neural spike identification

    Chaitanya Ekanadham, Daniel Tranchina, and Eero P Simoncelli. A blind sparse deconvolution method for neural spike identification. In Advances in Neural Information Processing Systems , pages 1440--1448, 2011

    2011

  17. [25]

    A Mathematical Introduction to Compressive Sensing

    Simon Foucart and Holger Rauhut. A Mathematical Introduction to Compressive Sensing . Springer, 2013

    2013

  18. [26]

    Fast online deconvolution of calcium imaging data

    Johannes Friedrich, Pengcheng Zhou, and Liam Paninski. Fast online deconvolution of calcium imaging data. PLoS computational biology , 13(3):e1005423, 2017

    2017

  19. [27]

    Toeplitz and circulant matrices: A review

    Robert M Gray et al. Toeplitz and circulant matrices: A review. Foundations and Trends in Communications and Information Theory , 2(3):155--239, 2006

    2006

  20. [28]

    Efficient dictionary learning with gradient descent

    Dar Gilboa, Sam Buchanan, and John Wright. Efficient dictionary learning with gradient descent. arXiv preprint arXiv:1809.10313 , 2018

    2018 arXiv

  21. [29]

    Cvx: Matlab software for disciplined convex programming, 2008

    Michael Grant, Stephen Boyd, and Yinyu Ye. Cvx: Matlab software for disciplined convex programming, 2008

    2008

  22. [30]

    Convolutional dictionary learning: A comparative review and new algorithms

    Cristina Garcia-Cardona and Brendt Wohlberg. Convolutional dictionary learning: A comparative review and new algorithms. IEEE Transactions on Computational Imaging , 4(3):366--381, 2018

    2018

  23. [31]

    On convergence rates of subgradient optimization methods

    Jean-Louis Goffin. On convergence rates of subgradient optimization methods. Mathematical programming , 13(1):329--347, 1977

    1977

  24. [32]

    Modeling regional and psychophysiologic interactions in fmri: the importance of hemodynamic deconvolution

    Darren R Gitelman, William D Penny, John Ashburner, and Karl J Friston. Modeling regional and psychophysiologic interactions in fmri: the importance of hemodynamic deconvolution. Neuroimage , 19(1):200--207, 2003

    2003

  25. [33]

    Implicit regularization in matrix factorization

    Suriya Gunasekar, Blake E Woodworth, Srinadh Bhojanapalli, Behnam Neyshabur, and Nati Srebro. Implicit regularization in matrix factorization. In Advances in Neural Information Processing Systems , pages 6151--6159, 2017

    2017

  26. [34]

    Fir perfect signal reconstruction from multiple convolutions: minimum deconvolver orders

    Gopal Harikumar and Yoram Bresler. Fir perfect signal reconstruction from multiple convolutions: minimum deconvolver orders. IEEE Transactions on Signal Processing , 46(1):215--218, 1998

    1998

  27. [35]

    Ultra-high resolution imaging by fluorescence photoactivation localization microscopy

    Samuel T Hess, Thanu PK Girirajan, and Michael D Mason. Ultra-high resolution imaging by fluorescence photoactivation localization microscopy. Biophysical journal , 91(11):4258--4272, 2006

    2006

  28. [36]

    Robust estimation of a location parameter

    Peter J Huber. Robust estimation of a location parameter. In Breakthroughs in statistics , pages 492--518. Springer, 1992

    1992

  29. [37]

    Daostorm: an algorithm for high-density super-resolution microscopy

    Seamus J Holden, Stephan Uphoff, and Achillefs N Kapanidis. Daostorm: an algorithm for high-density super-resolution microscopy. Nature methods , 8(4):279, 2011

    2011

  30. [38]

    Non-convex optimization for machine learning

    Prateek Jain, Purushottam Kar, et al. Non-convex optimization for machine learning. Foundations and Trends in Machine Learning , 10(3-4):142--336, 2017

    2017

  31. [39]

    Optimal injectivity conditions for bilinear inverse problems with applications to identifiability of deconvolution problems

    Michael Kech and Felix Krahmer. Optimal injectivity conditions for bilinear inverse problems with applications to identifiability of deconvolution problems. SIAM Journal on Applied Algebra and Geometry , 1(1):20--37, 2017

    2017

  32. [40]

    Geometry and symmetry in short-and-sparse deconvolution

    Han-Wen Kuo, Yenson Lau, Yuqian Zhang, and John Wright. Geometry and symmetry in short-and-sparse deconvolution. arXiv preprint arXiv:1901.00256 , 2019

    1901 arXiv

  33. [41]

    Suprema of chaos processes and the restricted isometry property

    Felix Krahmer, Shahar Mendelson, and Holger Rauhut. Suprema of chaos processes and the restricted isometry property. Communications on Pure and Applied Mathematics , 67(11):1877--1904, 2014

    1904

  34. [42]

    Multichannel blind deconvolution of seismic signals

    Kjetil F Kaaresen and Tofinn Taxt. Multichannel blind deconvolution of seismic signals. Geophysics , 63(6):2093--2107, 1998

    1998

  35. [43]

    Global geometry of multichannel sparse blind deconvolution on the sphere

    Yanjun Li and Yoram Bresler. Global geometry of multichannel sparse blind deconvolution on the sphere. arXiv preprint arXiv:1805.10437 , 2018

    2018 arXiv

  36. [44]

    Blind channel identification for speech dereverberation using l1-norm sparse learning

    Yuanqing Lin, Jingdong Chen, Youngmoo Kim, and Daniel D Lee. Blind channel identification for speech dereverberation using l1-norm sparse learning. In Advances in Neural Information Processing Systems , pages 921--928, 2008

    2008

  37. [45]

    Bilinear inverse problems with sparsity: optimal identifiability conditions and efficient recovery

    Yanjun Li. Bilinear inverse problems with sparsity: optimal identifiability conditions and efficient recovery . PhD thesis, University of Illinois at Urbana-Champaign, 2018

    2018

  38. [46]

    Spectral methods for passive imaging: Nonasymptotic performance and robustness

    Kiryung Lee, Felix Krahmer, and Justin Romberg. Spectral methods for passive imaging: Nonasymptotic performance and robustness. SIAM Journal on Imaging Sciences , 11(3):2110--2164, 2018

    2018

  39. [47]

    Identifiability in blind deconvolution with subspace or sparsity constraints

    Yanjun Li, Kiryung Lee, and Yoram Bresler. Identifiability in blind deconvolution with subspace or sparsity constraints. IEEE Transactions on Information Theory , 62(7):4266--4275, 2016

    2016

  40. [48]

    Identifiability in bilinear inverse problems with applications to subspace or sparsity-constrained blind gain and phase calibration

    Yanjun Li, Kiryung Lee, and Yoram Bresler. Identifiability in bilinear inverse problems with applications to subspace or sparsity-constrained blind gain and phase calibration. IEEE Transactions on Information Theory , 63(2):822--842, 2017

    2017

  41. [49]

    Blind gain and phase calibration via sparse spectral methods

    Yanjun Li, Kiryung Lee, and Yoram Bresler. Blind gain and phase calibration via sparse spectral methods. IEEE Transactions on Information Theory , 65(5):3097--3123, 2018

    2018

  42. [50]

    Rapid, robust, and reliable blind deconvolution via nonconvex optimization

    Xiaodong Li, Shuyang Ling, Thomas Strohmer, and Ke Wei. Rapid, robust, and reliable blind deconvolution via nonconvex optimization. Applied and Computational Harmonic Analysis , 2018

    2018

  43. [51]

    An overview of robust subspace recovery

    Gilad Lerman and Tyler Maunu. An overview of robust subspace recovery. Proceedings of the IEEE , 106(8):1380--1410, 2018

    2018

  44. [52]

    Short-and-sparse deconvolution -- a geometric approach

    Yenson Lau, Qing Qu, Han-Wen Kuo, Pengcheng Zhou, Yuqian Zhang, and John Wright. Short-and-sparse deconvolution -- a geometric approach. Preprint , 2019

    2019

  45. [53]

    Self-calibration and biconvex compressive sensing

    Shuyang Ling and Thomas Strohmer. Self-calibration and biconvex compressive sensing. Inverse Problems , 31(11):115002, 2015

    2015

  46. [54]

    Self-calibration and bilinear inverse problems via linear least squares

    Shuyang Ling and Thomas Strohmer. Self-calibration and bilinear inverse problems via linear least squares. SIAM Journal on Imaging Sciences , 11(1):252--292, 2018

    2018

  47. [55]

    Fast and guaranteed blind multichannel deconvolution under a bilinear system model

    Kiryung Lee, Ning Tian, and Justin Romberg. Fast and guaranteed blind multichannel deconvolution under a bilinear system model. IEEE Transactions on Information Theory , 64(7):4792--4818, 2018

    2018

  48. [56]

    Understanding blind deconvolution algorithms

    Anat Levin, Yair Weiss, Fredo Durand, and William T Freeman. Understanding blind deconvolution algorithms. IEEE Transactions on Pattern Analysis and Machine Intelligence , 33(12):2354--2367, 2011

    2011

  49. [57]

    Incremental methods for weakly convex optimization

    Xiao Li, Zhihui Zhu, Anthony Man-Cho So, and Jason D Lee. Incremental methods for weakly convex optimization. arXiv preprint arxiv.org:1907.11687 , 2019

    1907 arXiv

  50. [58]

    Nonconvex robust low-rank matrix recovery

    Xiao Li, Zhihui Zhu, Anthony Man-Cho So, and Rene Vidal. Nonconvex robust low-rank matrix recovery. arXiv preprint arXiv:1809.09237 , 2018

    2018 arXiv

  51. [59]

    Subspace methods for the blind identification of multichannel fir filters

    Eric Moulines, Pierre Duhamel, J-F Cardoso, and Sylvie Mayrargue. Subspace methods for the blind identification of multichannel fir filters. IEEE Transactions on signal processing , 43(2):516--525, 1995

    1995

  52. [60]

    Implicit regularization in nonconvex statistical estimation: Gradient descent converges linearly for phase retrieval, matrix completion and blind deconvolution

    Cong Ma, Kaizheng Wang, Yuejie Chi, and Yuxin Chen. Implicit regularization in nonconvex statistical estimation: Gradient descent converges linearly for phase retrieval, matrix completion and blind deconvolution. arXiv preprint arXiv:1711.10467 , 2017

    2017 arXiv

  53. [61]

    Sparse approximate solutions to linear systems

    Balas Kausik Natarajan. Sparse approximate solutions to linear systems. SIAM journal on computing , 24(2):227--234, 1995

    1995

  54. [62]

    A fast algorithm for sparse multichannel blind deconvolution

    Kenji Nose-Filho, Andr \'e K Takahata, Renato Lopes, and Jo \ a o MT Romano. A fast algorithm for sparse multichannel blind deconvolution. Geophysics , 81(1):V7--V16, 2015

    2015

  55. [63]

    Geometry of optimization and implicit regularization in deep learning

    Behnam Neyshabur, Ryota Tomioka, Ruslan Salakhutdinov, and Nathan Srebro. Geometry of optimization and implicit regularization in deep learning. arXiv preprint arXiv:1705.03071 , 2017

    2017 arXiv

  56. [64]

    Simultaneous denoising, deconvolution, and demixing of calcium imaging data

    Eftychios A Pnevmatikakis, Daniel Soudry, Yuanjun Gao, Timothy A Machado, Josh Merel, David Pfau, Thomas Reardon, Yu Mu, Clay Lacefield, Weijian Yang, et al. Simultaneous denoising, deconvolution, and demixing of calcium imaging data. Neuron , 89(2):285--299, 2016

    2016

  57. [65]

    Finding a sparse vector in a subspace: Linear sparsity using alternating directions

    Qing Qu, Ju Sun, and John Wright. Finding a sparse vector in a subspace: Linear sparsity using alternating directions. In Advances in Neural Information Processing Systems , pages 3401--3409, 2014

    2014

  58. [66]

    Convolutional phase retrieval

    Qing Qu, Yuqian Zhang, Yonina Eldar, and John Wright. Convolutional phase retrieval. In Advances in Neural Information Processing Systems , pages 6086--6096, 2017

    2017

  59. [67]

    Tsakiris, John Wright, and René Vidal

    Qing Qu, Zhihui Zhu, Xiao Li, Manolis C. Tsakiris, John Wright, and René Vidal. Finding the sparsest vectors in a subspace: Theory, algorithms, and applications, 2020

    2020

  60. [68]

    Sub-diffraction-limit imaging by stochastic optical reconstruction microscopy (storm)

    Michael J Rust, Mark Bates, and Xiaowei Zhuang. Sub-diffraction-limit imaging by stochastic optical reconstruction microscopy (storm). Nature methods , 3(10):793, 2006

    2006

  61. [69]

    Euclid in a taxicab: Sparse blind deconvolution with smoothed _1/ _2 regularization

    Audrey Repetti, Mai Quyen Pham, Laurent Duval, Emilie Chouzenoux, and Jean-Christophe Pesquet. Euclid in a taxicab: Sparse blind deconvolution with smoothed _1/ _2 regularization. IEEE signal processing letters , 22(5):539--543, 2015

    2015

  62. [70]

    Manifold gradient descent solves multi-channel sparse blind deconvolution provably and efficiently, 2019

    Laixi Shi and Yuejie Chi. Manifold gradient descent solves multi-channel sparse blind deconvolution provably and efficiently, 2019

    2019

  63. [71]

    Image reconstruction from phased-array data based on multichannel blind deconvolution

    Huajun She, Rong-Rong Chen, Dong Liang, Yuchou Chang, and Leslie Ying. Image reconstruction from phased-array data based on multichannel blind deconvolution. Magnetic resonance imaging , 33(9):1106--1113, 2015

    2015

  64. [72]

    Robust multichannel blind deconvolution via fast alternating minimization

    Filip Sroubek and Peyman Milanfar. Robust multichannel blind deconvolution via fast alternating minimization. IEEE Transactions on Image processing , 21(4):1687--1700, 2012

    2012

  65. [73]

    Deconvolution methods for 3-d fluorescence microscopy images

    Pinaki Sarder and Arye Nehorai. Deconvolution methods for 3-d fluorescence microscopy images. IEEE Signal Processing Magazine , 23(3):32--45, 2006

    2006

  66. [74]

    Complete dictionary recovery over the sphere i: Overview and the geometric picture

    Ju Sun, Qing Qu, and John Wright. Complete dictionary recovery over the sphere i: Overview and the geometric picture. IEEE Transactions on Information Theory , 63(2):853--884, 2016

    2016

  67. [75]

    Complete dictionary recovery over the sphere ii: Recovery by riemannian trust-region method

    Ju Sun, Qing Qu, and John Wright. Complete dictionary recovery over the sphere ii: Recovery by riemannian trust-region method. IEEE Transactions on Information Theory , 63(2):885--914, 2017

    2017

  68. [76]

    Fluorophore localization algorithms for super-resolution microscopy

    Alex Small and Shane Stahlheber. Fluorophore localization algorithms for super-resolution microscopy. Nature methods , 11(3):267, 2014

    2014

  69. [77]

    Provable nonconvex methods/algorithms

    Ju Sun. Provable nonconvex methods/algorithms. https://sunju.org/research/nonconvex/

  70. [78]

    Upper and lower bounds for stochastic processes: modern methods and classical problems , volume 60

    Michel Talagrand. Upper and lower bounds for stochastic processes: modern methods and classical problems , volume 60. Springer Science & Business Media, 2014

    2014

  71. [79]

    Multichannel myopic deconvolution in underwater acoustic channels via low-rank recovery

    Ning Tian, Sung-Hoon Byun, Karim Sabra, and Justin Romberg. Multichannel myopic deconvolution in underwater acoustic channels via low-rank recovery. The Journal of the Acoustical Society of America , 141(5):3337--3348, 2017

    2017

  72. [80]

    Dual principal component pursuit

    Manolis C Tsakiris and Ren \'e Vidal. Dual principal component pursuit. In Proceedings of the IEEE International Conference on Computer Vision Workshops , pages 10--18, 2015

    2015

  73. [81]

    Blind deconvolution from multiple sparse inputs

    Liming Wang and Yuejie Chi. Blind deconvolution from multiple sparse inputs. IEEE Signal Processing Letters , 23(10):1384--1388, 2016

    2016

  74. [82]

    A blind deconvolution approach to recover effective connectivity brain networks from resting state fmri data

    Guo-Rong Wu, Wei Liao, Sebastiano Stramaglia, Ju-Rong Ding, Huafu Chen, and Daniele Marinazzo. A blind deconvolution approach to recover effective connectivity brain networks from resting state fmri data. Medical image analysis , 17(3):365--374, 2013

    2013

  75. [83]

    A least-squares approach to blind channel identification

    Guanghan Xu, Hui Liu, Lang Tong, and Thomas Kailath. A least-squares approach to blind channel identification. IEEE Transactions on signal processing , 43(12):2982--2993, 1995

    1995

  76. [84]

    Structured local minima in sparse blind deconvolution

    Yuqian Zhang, Han-wen Kuo, and John Wright. Structured local minima in sparse blind deconvolution. In S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, and R. Garnett, editors, Advances in Neural Information Processing Systems 31 , pages 2328--2337. Curran Ass...

    2018

  77. [85]

    On the global geometry of sphere-constrained sparse blind deconvolution

    Yuqian Zhang, Yenson Lau, Han-Wen Kuo, Sky Cheung, Abhay Pasupathy, and John Wright. On the global geometry of sphere-constrained sparse blind deconvolution. In Computer Vision and Pattern Recognition (CVPR), 2017 IEEE Conference on , pages 4381--4389. IEEE, 2017

    2017

  78. [86]

    Dual principal component pursuit: Improved analysis and efficient algorithms

    Zhihui Zhu, Yifan Wang, Daniel Robinson, Daniel Naiman, Rene Vidal, and Manolis Tsakiris. Dual principal component pursuit: Improved analysis and efficient algorithms. In Advances in Neural Information Processing Systems , pages 2171--2181, 2018

    2018

  79. [87]

    Multi-image blind deblurring using a coupled adaptive sparse prior

    Haichao Zhang, David Wipf, and Yanning Zhang. Multi-image blind deblurring using a coupled adaptive sparse prior. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages 1051--1058, 2013

    2013

  80. [88]

    Complete dictionary learning via ^ 4 -norm maximization over the orthogonal group

    Yuexiang Zhai, Zitong Yang, Zhenyu Liao, John Wright, and Yi Ma. Complete dictionary learning via ^ 4 -norm maximization over the orthogonal group. arXiv preprint arXiv:1906.02435 , 2019

    1906 arXiv

  81. [89]

    Faster storm using compressed sensing

    Lei Zhu, Wei Zhang, Daniel Elnatan, and Bo Huang. Faster storm using compressed sensing. Nature methods , 9(7):721, 2012

    2012

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.