MAP Image Recovery with Guarantees using Locally Convex Multi-Scale Energy (LC-MUSE) Model

Jyothi Rikhab Chand; Mathews Jacob

REVIEW 3 major objections 6 minor 25 references

MAP Image Recovery with Guarantees using Locally Convex Multi-Scale Energy (LC-MUSE) Model

T0 review · 3 major / 6 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read A locally convex deep energy model gives MAP image recovery a unique minimizer, convergent optimization, and a noise-robustness bound, while matching state-of-the-art MRI reconstruction quality.

desk verdict Local-convexity training scheme is a genuine advance, but the advertised guarantees rest on conditions that are never verified on the trained model; the empirical comparison is solid. read the letter →

arxiv 2502.03302 v1 pith:UYJ7IU6L submitted 2025-02-05 cs.LG cs.CVeess.IV

Jyothi Rikhab Chand , Mathews Jacob This is my paper · ORCID

classification cs.LGcs.CVeess.IV MSC 68T0768U1094A0865K10

keywords locallyconvexenergymodelMAPestimationinverseproblemsparallelMRIreconstructionscorematchingmonotoneoperatorLipschitzconstraintdeeplearningregularizer

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 aims to give deep-learning-based maximum a posteriori (MAP) image recovery the same guarantees that classical convex reconstruction enjoys, without giving up the image quality of modern learned priors. Its proposal, LC-MuSE, is a multi-scale energy model whose negative log-prior is strongly convex only inside a ball around the data manifold, a much weaker condition than global convexity. The authors prove that with this condition the MAP objective has a unique minimizer in the ball, the MM optimization converges from any initialization in the ball, and the reconstruction is stable to measurement perturbations. On parallel fastMRI brain reconstruction, LC-MuSE reports better PSNR and SSIM than the globally convex ICNN baseline and performance comparable to the non-convex MuSE and end-to-end MoL methods. If these properties hold, learned image reconstruction can be both high-quality and certified.

What carries the argument

The load-bearing object is the locally monotone score map. The prior is an explicit energy $E_\theta(x)=\frac{1}{2\sigma_f^2}\|x-\Psi_\theta(x)\|^2$ with score $H_\theta=\nabla_x E_\theta$; by Lemma 2.1, strong convexity of $E_\theta$ is equivalent to strong monotonicity of $H_\theta$, and Lemma 3.1 extends this to a ball $B_\delta(u)$, giving $m$-strong local convexity. The training loss (9) adds a ReLU penalty on $L[T_\theta(x)]-l$, where $L[T_\theta(x)]$ is the local Lipschitz constant of $T_\theta=I-H_\theta$, estimated by projected gradient ascent inside $B_\delta(x)$. These components make the MAP objective locally convex, and the MM update (11) then alternates a linear solve with a gradient step of the energy to minimize it.

What would settle it

On a held-out test image, compute the actual local Lipschitz constant $\max_{x_1,x_2\in B_\delta(x^*)} \|T_\theta(x_1)-T_\theta(x_2)\|/\|x_1-x_2\|$; if it exceeds $1-m$, or if $\|x_0-x^*\|>\delta$ for the SENSE initialization, then Lemma 3.4's premise is violated and the stated convergence guarantee need not hold.

Watch

Extended reading notes

Core claim

The central claim is that one can enforce local strong convexity of a deep energy prior by constraining its score function to be locally monotone, and that this local version of convexity preserves the guarantees that global convexity is normally required for. Specifically, the energy $E_\theta(x)=\frac{1}{2\sigma_f^2}\|x-\Psi_\theta(x)\|^2$ is trained with a multi-scale score-matching loss plus a ReLU penalty on $L[T_\theta(x)]$, the local Lipschitz constant of $T_\theta=I-\nabla E_\theta$, forcing $L[T_\theta(x)]\le l=1-m$ on balls around training points. With the resulting $m$-strong locally convex energy in the MAP objective $f(x,b)$, the paper proves uniqueness of the minimizer within the ball (Lemma 3.3), convergence of the MM iterates from any initialization in the ball (Lemma 3.4), and the perturbation bound $\|x^*(b+n)-x^*(b)\|\le\|n\|/(m\eta^2)$ (Lemma 3.5). The authors also report that on fastMRI brain data, LC-MuSE reaches higher PSNR/SSIM than the globally convex ICNN baseline and performance comparable to MuSE and E2E-MoL.

Load-bearing premise

The guarantees hold only if the trained network really is locally monotone on the ball around each test image, but the training loss only penalizes violations with finite weight and never verifies the constraint, and the paper never checks that SENSE initializations on test images lie inside the chosen ball.

Editorial extensions

If this is right

Within the local convexity ball, the MAP minimizer is unique, so the reconstruction cannot depend on which of several equally good solutions the optimizer happens to find.
Initializing with SENSE places the algorithm in the basin of attraction, so the MM iterates converge to the unique minimizer rather than stalling at spurious local minima.
The bound $\|x^*(b+n)-x^*(b)\|\le\|n\|/(m\eta^2)$ quantifies robustness: a measurement perturbation changes the recovery by at most a constant times the perturbation size.
Since the constraint is local and not architectural, arbitrary CNN backbones can be trained as regularizers, avoiding the quality loss reported for the globally convex ICNN.
The same framework carries over to any linear inverse problem with a reliable initializer, not only parallel MRI.

Reading between the lines

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

Implicitly, the soft penalty in (9) means the stated guarantees are conditional on a constraint that is never verified; a test-time check of the local Lipschitz bound would turn this into a certified pipeline.
The robustness bound suggests raising $m$ improves noise stability, but doing so tightens the required Lipschitz bound and likely trades away reconstruction quality, so $m$ is a practical stability-versus-fidelity knob.
The same local-convexity construction could be adapted to nonlinear forward operators or to posterior sampling, where local strong convexity would control the posterior landscape rather than only the MAP point.
The choice of $\delta$ as the worst-case SENSE error on training data is heuristic; per-test-image radius selection would make the convergence guarantee reliable outside the training distribution.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

The useful new thing here is the LC-MuSE training scheme: instead of forcing global convexity through architecture (ICNN) or global Lipschitzness, it penalizes violations of local monotonicity with an adversarial loss. That is a sensible way to get some of the benefits of convexity without the performance penalty, and the multi-scale score-matching base is a reasonable foundation. The experiments back up the performance claims: LC-MuSE clearly beats ICNN, and matches MuSE and E2E-MoL on the fastMRI brain data. The numbers look credible, and the figures support the stated comparisons.

The soft spots are real but addressable. First, Lemmas 3.1 through 3.5 are stated without proofs; Lemma 3.1 even says the proof is omitted. For a theory paper that is a problem, though these are plausible local extensions of known monotone-operator results. Second, and more important, the paper never checks that the trained network actually satisfies the local monotonicity condition. Loss (9) is a soft ReLU penalty with finite λ, and the local Lipschitz constant in (10) is estimated by projected gradient ascent, which may not find the true maximum. A post-training measurement of L[T_θ(x)] on held-out data would close this gap. Third, the choice of δ is off. In Section IV-B, δ is set to the worst-case deviation of SENSE reconstructions from reference images in the training set. But Lemma 3.4 requires ||x0 - x*|| ≤ δ, where x* is the minimizer of the MAP objective, not the reference image. SENSE being close to the reference does not imply closeness to the MAP solution, and the distance from SENSE initializations to converged MAP solutions on test data is never reported. Without that check, Lemmas 3.3–3.5 do not actually apply to the reconstructions shown.

That said, the central idea is sound and the empirical work is honest. The paper is a genuine step toward practical locally convex priors, and the gaps are fixable: add proofs (or at least sketches) for the lemmas, report achieved local Lipschitz constants on test data, and verify the initialization condition empirically. If those are done, the contribution would be solid. This paper deserves a serious referee, not a desk rejection, but the referee should push for those fixes. I would bring it to a reading group for the method itself, and I would cite it as related work if the authors follow through on the verification.

Referee Report

3 major / 6 minor

Summary. The manuscript introduces LC-MuSE, a multi-scale energy model trained so that its associated negative log-prior is m-strongly locally convex in a ball around the image manifold. The training objective (9) combines score matching with a soft penalty on the local Lipschitz constant of T_theta = I - grad E_theta. The authors claim that the resulting MAP objective (1) has a unique minimizer, that the MM updates (11) converge from SENSE initializations, and that the minimizer obeys the perturbation bound (12). The method is evaluated on fastMRI brain data, reporting PSNR/SSIM improvements over ICNN and results comparable to MuSE and E2E-MoL.

Significance. If the theoretical claims were fully established, the paper would make a valuable contribution: it relaxes the architectural restrictions of input-convex networks while retaining local strong-convexity-style guarantees, and it offers an interesting demonstration that a locally convex energy can match end-to-end trained methods in MR reconstruction. The empirical comparison is a genuine strength: the method is evaluated on a public dataset against several state-of-the-art baselines, and the reported numbers support the comparative claims. However, the central 'guarantees' are not substantiated as written: the lemmas underlying uniqueness, convergence, and robustness are unproved, the trained model is not shown to satisfy the required local monotonicity condition, and the calibration of delta is based on the wrong distance. These gaps must be closed before the guarantees can be accepted.

major comments (3)

[Section III-C] Lemmas 3.1 and 3.2 are dismissed with 'We omit the proof because of space constraints,' and Lemmas 3.3, 3.4, and 3.5 are stated without proof or attribution to a specific theorem in the cited prior work [5,7]. These lemmas are the entire basis for the paper's claims of uniqueness, convergence, and robustness. In particular, the perturbation bound (12) and the accompanying condition ||n|| <= m*delta*eta^2 are nontrivial: they require showing that the perturbed minimizer remains in the ball B_delta(x*) and carefully tracking the strong convexity modulus of the total objective (1). The current text does not provide this derivation. For a journal submission, complete proofs or precise references to existing theorems must be supplied.
[Section III-A, Eq. (9)] The training loss adds a finite-weight ReLU penalty, lambda * R(L[T_theta(x)] - l)^2, which does not enforce the local Lipschitz constraint (8) exactly. Lemmas 3.2-3.5 require L[T_theta(x)] <= l = 1 - m to hold, at least on the domain of interest. The paper reports no post-training measurement of the achieved local Lipschitz constant on either training or test data. Moreover, the estimate in (10) is obtained by projected gradient ascent, which may converge to a local maximum rather than the true maximum of the local Lipschitz ratio; hence even the training-time quantity is not certified. Without verification after training, the theoretical guarantees do not apply to the LC-MuSE model whose reconstructions are reported in Section IV.
[Section IV-B] The radius delta is chosen as the worst-case deviation of SENSE reconstructions from reference images in the training set. However, the hypothesis required by Lemma 3.4 is ||x_0 - x*|| <= delta, where x* is the minimizer of (1), not the reference image. SENSE-to-reference closeness does not imply SENSE-to-MAP closeness, and the paper never checks the distance from SENSE initializations to the converged MAP solutions on test images. Therefore, the convergence guarantee stated in Lemma 3.4 is not connected to the initialization strategy used in the experiments. The authors should either calibrate delta using ||SENSE - x*|| or report empirical values of this distance on the test set.

minor comments (6)

[Section IV-D and Fig. 1] The text states that Fig. 1a and 1b show reconstructions for two-fold and four-fold accelerations, respectively, but the caption of Fig. 1b says six-fold acceleration; please correct this inconsistency.
[Section III-B, Eq. (11)] The scalar L in the update rule is not defined before first use; please specify whether it is the Lipschitz constant of grad E_theta and how it relates to the parameter l in (8).
[Section III-C] The statement of Lemma 3.3, 'the solution of (1) is unique within B_delta(x)', is ambiguous because the same symbol x is used for the center of the ball and for the optimization variable; the center should be identified explicitly, e.g., as x*.
[Section III-A, Eq. (10)] The local Lipschitz ratio is undefined when x_1 = x_2; the maximum should be taken over distinct points in B_delta(x).
[Section IV-B] The numerical values of delta, m, lambda, and the range of sigma_f used in training are not reported; these hyperparameters control the size of the ball and the validity of Lemma 3.5, so they should be listed.
[Section II, Eq. (2)] Equation (2) defines E_theta with the factor 1/(2 sigma_f^2), but the role of sigma_f is not explained until the update rule (11); please clarify the relationship between sigma_f and the noise scales used in score matching.

Circularity Check

0 steps flagged · score 0.0 of 10

No load-bearing circularity; the central contribution is a new training scheme, and the theoretical guarantees are standard consequences of local strong convexity rather than reductions to fitted outputs.

full rationale

The paper's derivation chain does not exhibit any step in which a predicted result is equivalent by construction to an input, fitted parameter, or self-citation. The training objective (9) combines score matching Jθ with a soft ReLU penalty on the local Lipschitz estimate L[Tθ(x)]; the subsequent lemmas (3.3–3.5) are conditional statements that would follow from m-strong local convexity if the constraint were exactly met. No equation in the paper reduces Lemma 3.3, 3.4, or 3.5 to the training loss itself. The self-citations to [5] and [7] supply a standard monotone-operator lemma and an MM update rule, respectively; both are external, published results with stated assumptions, so they are not circular. The paper does contain genuine gaps: the local monotonicity constraint is only encouraged with finite λ, never verified after training; Lemma 3.1's proof is omitted; and δ in Section IV-B is calibrated as the worst-case SENSE-to-reference deviation, while Lemma 3.4 requires ∥x0−x*∥≤δ with x* the MAP minimizer. These are unverified premises and empirical mismatches, not circular reductions. The comparison against ICNN, MuSE, and E2E-MoL on fastMRI is an external benchmark, so the performance claim is not forced by construction. Accordingly, the circularity score is 0.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central method depends on four quantities that are either tuned (λ, δ) or implied (m, σf) and never reported numerically. The main theoretical assumption is that the soft penalty in (9) actually brings the trained network into the local Lipschitz regime required by the lemmas, plus the domain assumption that SENSE initialization stays inside the local convexity ball at test time.

free parameters (4)

lambda (penalty weight)
Controls how exactly the local Lipschitz constraint L[Tθ(x)] ≤ l is enforced; no value or schedule given, and the final constraint satisfaction is not reported.
delta (local convexity radius) = not reported numerically (worst-case SENSE deviation on training set)
Defines the ball Bδ(x) where strong convexity holds; chosen from training data, not verified for test images.
m (strong convexity modulus)
Appears in the local monotonicity condition and robustness bound; the paper never states what m or l = 1−m is achieved by the trained model.
sigma_f (finest noise scale)
Used in the update rule (11) via ζ = η/σf; must be set to match the training noise level.

assumptions (4)

standard math Local strong convexity of Eθ is equivalent to local strong monotonicity of its gradient (Lemma 3.1)
Extension of classical convex analysis; stated without proof, but follows from standard results if the ball is convex.
standard math A Lipschitz bound L[Tθ] ≤ 1−m on a ball implies m-strong local monotonicity of Hθ (Lemma 3.2)
Local version of Lemma 2.2 from [5]; again no proof given.
ad hoc to paper The penalty in (9) with finite λ effectively enforces the local Lipschitz constraint on test images
The paper treats the soft penalty as if it guarantees the condition needed for Lemmas 3.3 to 3.5, but does not verify it after training.
domain assumption SENSE initialization lies within Bδ(x∗) for every test image
δ is chosen from training-set worst-case SENSE error, then assumed to transfer to the test set without validation.

how reviews work

0 comments

Cite this review

Pith. "Pith review of MAP Image Recovery with Guarantees using Locally Convex Multi-Scale Energy (LC-MUSE) Model." pith.science (2026). https://pith.science/paper/UYJ7IU6L

@misc{pith2026250203302,
  author       = {Pith},
  title        = {Pith review of: MAP Image Recovery with Guarantees using Locally Convex Multi-Scale Energy (LC-MUSE) Model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UYJ7IU6L}},
  note         = {Machine review of arXiv:2502.03302}
}

read the original abstract

We propose a multi-scale deep energy model that is strongly convex in the local neighbourhood around the data manifold to represent its probability density, with application in inverse problems. In particular, we represent the negative log-prior as a multi-scale energy model parameterized by a Convolutional Neural Network (CNN). We restrict the gradient of the CNN to be locally monotone, which constrains the model as a Locally Convex Multi-Scale Energy (LC-MuSE). We use the learned energy model in image-based inverse problems, where the formulation offers several desirable properties: i) uniqueness of the solution, ii) convergence guarantees to a minimum of the inverse problem, and iii) robustness to input perturbations. In the context of parallel Magnetic Resonance (MR) image reconstruction, we show that the proposed method performs better than the state-of-the-art convex regularizers, while the performance is comparable to plug-and-play regularizers and end-to-end trained methods.

Figures

Figures reproduced from arXiv: 2502.03302 by the authors.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

25 extracted references · 17 canonical work pages

[1]

A. N. Tikhonov and V . I. Arsenin, Solutions of Ill-posed Problems: Andrey N. Tikhonov and Vasiliy Y. Arsenin. Translation Editor Fritz John. Wiley, 1977

work page 1977
[2]

Plug-and-play priors for model based reconstruction,

S. V . Venkatakrishnan, C. A. Bouman, and B. Wohlberg, “Plug-and-play priors for model based reconstruction,” in 2013 IEEE Global Conference on Signal and Information Processing , 2013, pp. 945–948

work page 2013
[3]

Plug- and-play unplugged: Optimization-free reconstruction using consensus equilibrium,

G. T. Buzzard, S. H. Chan, S. Sreehari, and C. A. Bouman, “Plug- and-play unplugged: Optimization-free reconstruction using consensus equilibrium,” SIAM Journal on Imaging Sciences , vol. 11, no. 3, pp. 2001–2020, 2018

work page 2001
[4]

Plug-and-play methods for magnetic res- onance imaging: Using denoisers for image recovery,

R. Ahmad, C. A. Bouman, G. T. Buzzard, S. Chan, S. Liu, E. T. Reehorst, and P. Schniter, “Plug-and-play methods for magnetic res- onance imaging: Using denoisers for image recovery,” IEEE signal processing magazine, vol. 37, no. 1, pp. 105–116, 2020

work page 2020
[5]

Memory-efficient model-based deep learning with convergence and robustness guarantees,

A. Pramanik, M. B. Zimmerman, and M. Jacob, “Memory-efficient model-based deep learning with convergence and robustness guarantees,” IEEE Transactions on Computational Imaging , vol. 9, pp. 260–275, 2023

work page 2023
[6]

Plug-and- play methods provably converge with properly trained denoisers,

E. Ryu, J. Liu, S. Wang, X. Chen, Z. Wang, and W. Yin, “Plug-and- play methods provably converge with properly trained denoisers,” in International Conference on Machine Learning . PMLR, 2019, pp. 5546–5557

work page 2019
[7]

Multi-scale energy (muse) framework for inverse problems in imaging,

J. R. Chand and M. Jacob, “Multi-scale energy (muse) framework for inverse problems in imaging,” IEEE Transactions on Computational Imaging, pp. 1–16, 2024

work page 2024
[8]

Deep admm-net for compressive sensing mri,

J. Sun, H. Li, Z. Xu et al. , “Deep admm-net for compressive sensing mri,” Advances in neural information processing systems , vol. 29, 2016

2016

Show all 25 references

[9]

Learning a variational network for reconstruction of accelerated mri data,

K. Hammernik, T. Klatzer, E. Kobler, M. P. Recht, D. K. Sodickson, T. Pock, and F. Knoll, “Learning a variational network for reconstruction of accelerated mri data,” Magnetic resonance in medicine, vol. 79, no. 6, pp. 3055–3071, 2018

2018
[10]

Modl: Model-based deep learning architecture for inverse problems,

H. K. Aggarwal, M. P. Mani, and M. Jacob, “Modl: Model-based deep learning architecture for inverse problems,” IEEE transactions on medical imaging, vol. 38, no. 2, pp. 394–405, 2018

2018
[11]

Cinenet: deep learning-based 3d cardiac cine mri reconstruction with multi- coil complex-valued 4d spatio-temporal convolutions,

T. K ¨ustner, N. Fuin, K. Hammernik, A. Bustin, H. Qi, R. Hajhosseiny, P. G. Masci, R. Neji, D. Rueckert, R. M. Botnar et al. , “Cinenet: deep learning-based 3d cardiac cine mri reconstruction with multi- coil complex-valued 4d spatio-temporal convolutions,”Scientific reports,...

2020
[12]

It has potential: Gradient-driven denoisers for convergent solutions to inverse problems,

R. Cohen, Y . Blau, D. Freedman, and E. Rivlin, “It has potential: Gradient-driven denoisers for convergent solutions to inverse problems,” Advances in Neural Information Processing Systems , vol. 34, pp. 18 152–18 164, 2021

2021
[13]

Gradient step denoiser for convergent plug-and-play,

S. Hurault, A. Leclaire, and N. Papadakis, “Gradient step denoiser for convergent plug-and-play,” arXiv preprint arXiv:2110.03220 , 2021

2021 arXiv
[14]

Learned convex regularizers for inverse problems,

S. Mukherjee, S. Dittmer, Z. Shumaylov, S. Lunz, O. ¨Oktem, and C.-B. Sch¨onlieb, “Learned convex regularizers for inverse problems,” arXiv preprint arXiv:2008.02839, 2020

2008 arXiv
[15]

A neural-network-based convex regularizer for inverse problems,

A. Goujon, S. Neumayer, P. Bohra, S. Ducotterd, and M. Unser, “A neural-network-based convex regularizer for inverse problems,” IEEE Transactions on Computational Imaging , 2023

2023
[16]

Learning weakly convex regu- larizers for convergent image-reconstruction algorithms,

A. Goujon, S. Neumayer, and M. Unser, “Learning weakly convex regu- larizers for convergent image-reconstruction algorithms,” SIAM Journal on Imaging Sciences , vol. 17, no. 1, pp. 91–115, 2024

2024
[17]

Weakly convex regularisers for inverse problems: Convergence of critical points and primal-dual optimisation,

Z. Shumaylov, J. Budd, S. Mukherjee, and C.-B. Sch ¨onlieb, “Weakly convex regularisers for inverse problems: Convergence of critical points and primal-dual optimisation,” arXiv preprint arXiv:2402.01052 , 2024

2024 arXiv
[18]

A connection between score matching and denoising au- toencoders,

P. Vincent, “A connection between score matching and denoising au- toencoders,” Neural computation, vol. 23, no. 7, pp. 1661–1674, 2011

2011
[19]

E. K. Ryu and W. Yin, Large-scale convex optimization: algorithms & analyses via monotone operators . Cambridge University Press, 2022

2022
[20]

Spectral normalization for generative adversarial networks,

T. Miyato, T. Kataoka, M. Koyama, and Y . Yoshida, “Spectral normalization for generative adversarial networks,” arXiv preprint arXiv:1802.05957, 2018

2018 arXiv
[21]

Clip: Cheap lipschitz training of neural networks,

L. Bungert, R. Raab, T. Roith, L. Schwinn, and D. Tenbrinck, “Clip: Cheap lipschitz training of neural networks,” inInternational Conference on Scale Space and Variational Methods in Computer Vision. Springer, 2021, pp. 307–319

2021
[22]

Espirit—an eigenvalue approach to autocalibrating parallel mri: where sense meets grappa,

M. Uecker, P. Lai, M. J. Murphy, P. Virtue, M. Elad, J. M. Pauly, S. S. Vasanawala, and M. Lustig, “Espirit—an eigenvalue approach to autocalibrating parallel mri: where sense meets grappa,” Magnetic resonance in medicine , vol. 71, no. 3, pp. 990–1001, 2014

2014
[23]

fastmri: An open dataset and benchmarks for accelerated mri,

J. Zbontar, F. Knoll, A. Sriram, T. Murrell, Z. Huang, M. J. Muck- ley, A. Defazio, R. Stern, P. Johnson, M. Bruno et al. , “fastmri: An open dataset and benchmarks for accelerated mri,” arXiv preprint arXiv:1811.08839, 2018

2018 arXiv
[24]

Sense: sensitivity encoding for fast mri,

K. P. Pruessmann, M. Weiger, M. B. Scheidegger, and P. Boesiger, “Sense: sensitivity encoding for fast mri,” Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine , vol. 42, no. 5, pp. 952–962, 1999

1999
[25]

Deep equilibrium models,

S. Bai, J. Z. Kolter, and V . Koltun, “Deep equilibrium models,”Advances in Neural Information Processing Systems , vol. 32, 2019

2019

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.

[1] [1]

A. N. Tikhonov and V . I. Arsenin, Solutions of Ill-posed Problems: Andrey N. Tikhonov and Vasiliy Y. Arsenin. Translation Editor Fritz John. Wiley, 1977

work page 1977

[2] [2]

Plug-and-play priors for model based reconstruction,

S. V . Venkatakrishnan, C. A. Bouman, and B. Wohlberg, “Plug-and-play priors for model based reconstruction,” in 2013 IEEE Global Conference on Signal and Information Processing , 2013, pp. 945–948

work page 2013

[3] [3]

Plug- and-play unplugged: Optimization-free reconstruction using consensus equilibrium,

G. T. Buzzard, S. H. Chan, S. Sreehari, and C. A. Bouman, “Plug- and-play unplugged: Optimization-free reconstruction using consensus equilibrium,” SIAM Journal on Imaging Sciences , vol. 11, no. 3, pp. 2001–2020, 2018

work page 2001

[4] [4]

Plug-and-play methods for magnetic res- onance imaging: Using denoisers for image recovery,

R. Ahmad, C. A. Bouman, G. T. Buzzard, S. Chan, S. Liu, E. T. Reehorst, and P. Schniter, “Plug-and-play methods for magnetic res- onance imaging: Using denoisers for image recovery,” IEEE signal processing magazine, vol. 37, no. 1, pp. 105–116, 2020

work page 2020

[5] [5]

Memory-efficient model-based deep learning with convergence and robustness guarantees,

A. Pramanik, M. B. Zimmerman, and M. Jacob, “Memory-efficient model-based deep learning with convergence and robustness guarantees,” IEEE Transactions on Computational Imaging , vol. 9, pp. 260–275, 2023

work page 2023

[6] [6]

Plug-and- play methods provably converge with properly trained denoisers,

E. Ryu, J. Liu, S. Wang, X. Chen, Z. Wang, and W. Yin, “Plug-and- play methods provably converge with properly trained denoisers,” in International Conference on Machine Learning . PMLR, 2019, pp. 5546–5557

work page 2019

[7] [7]

Multi-scale energy (muse) framework for inverse problems in imaging,

J. R. Chand and M. Jacob, “Multi-scale energy (muse) framework for inverse problems in imaging,” IEEE Transactions on Computational Imaging, pp. 1–16, 2024

work page 2024

[8] [8]

Deep admm-net for compressive sensing mri,

J. Sun, H. Li, Z. Xu et al. , “Deep admm-net for compressive sensing mri,” Advances in neural information processing systems , vol. 29, 2016

2016