PtyGenography: using generative models for regularization of the phase retrieval problem

Allard Mosk; Palina Salanevich; Selin Aslan; Tristan van Leeuwen

REVIEW 4 major objections 6 minor 1 cited by

PtyGenography: using generative models for regularization of the phase retrieval problem

T0 review · 4 major / 6 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read A unified variational estimator interpolates between classical and generative phase retrieval, with reconstruction error bounded by the noise level when the coupling parameter tracks the noise.

desk verdict A clean idea and honest writing, but the experiment doesn't test the proposed method and the theory doesn't actually show the bridging claim. read the letter →

arxiv 2502.01338 v1 pith:MD7TYNVM submitted 2025-02-03 stat.ML cs.ITcs.LGmath.FAmath.ITmath.OC

Selin Aslan , Tristan van Leeuwen , Allard Mosk , Palina Salanevich This is my paper · ORCID

classification stat.MLcs.ITcs.LGmath.FAmath.ITmath.OC MSC 65J2268T0794A12

keywords phaseretrievalgenerativepriorsregularizationinverseproblemsmaskedFouriermeasurementsbias-variancetrade-offvariationalreconstructionnoiserobustness

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 studies two ways to reconstruct a signal from noisy intensity measurements: minimizing the measurement error directly, or minimizing it over the latent space of a generative model whose outputs form the allowed set of signals. The direct method amplifies noise; the generative method suppresses noise but introduces a bias when the true signal is not exactly in the model's range. The paper derives reconstruction-error bounds for both methods in terms of noise and model bias, and then proposes a unified variational objective that combines the two. If the coupling parameter is chosen proportional to the noise level, the unified estimator's error is bounded by a constant times the noise level. Numerical experiments on masked Fourier measurements of handwritten digits show the unified method matching or beating both extremes across signal-to-noise ratios.

What carries the argument

The load-bearing object is the coupled variational problem (3), written as $\min_{z,f}\|A(f)-y\|_2^2+\lambda^2\|G(z)-f\|_2^2$; it interpolates continuously between the classical formulation (1) ($\lambda=0$) and the generative formulation (2) ($\lambda\to\infty$) by pairing the measurement fidelity term with a quadratic penalty that pulls the reconstructed signal toward the generative model's range. The proofs ride on bi-Lipschitz constants: $\alpha$ for the measurement map $A$, $\beta$ for the generator $G$, and $\gamma$ for the composition $A\circ G$. The generator is claimed to regularize the problem precisely when $\gamma<\alpha$, and the bias term $\|G(z_0)-f_0\|_2$ measures how far the true signal lies outside the model range. The numerical experiments optimize the unified objective in an equivalent form $\min_x\|A\circ B(x)-y\|_2^2+\lambda^2\|w\odot x\|_2^2$ using L-BFGS.

What would settle it

Compute the effective stretch factors $\gamma$ and $\alpha$ for the masked Fourier operator and the PCA digits generator by evaluating $\|A\circ G(z)-A\circ G(z')\|/\|z-z'\|$ and $\|A(f)-A(f')\|/\|f-f'\|$ over many pairs; if the smallest such factor for $A\circ G$ is not below that for $A$, the Lemma 2 noise-suppression advantage disappears and the theoretical rationale for the unified method is undermined.

Watch

Extended reading notes

Core claim

The central claim is that the bias-variance trade-off between classical and generative phase retrieval can be described by explicit bounds and exploited constructively. For the classical problem the paper obtains $\|\tilde f-f_0\|_2\le 2\alpha\|\varepsilon\|_2$; for the generative problem it obtains $\|\tilde f-f_0\|_2\le (1+2\alpha\beta\gamma)\|G(z_0)-f_0\|_2+2\beta\gamma\|\varepsilon\|_2$, so noise amplification is governed by the composition constant $\gamma$ while a model bias $G(z_0)-f_0$ enters. The proposed unified estimator $\min_{z,f}\|A(f)-y\|_2^2+\lambda^2\|G(z)-f\|_2^2$ recovers the two extremes at $\lambda=0$ and $\lambda\to\infty$, and Lemma 3 gives $\|\tilde f-f_0\|_2\le \lambda\alpha\|G(z_0)-f_0\|_2+2\alpha\|\varepsilon\|_2$. With $\lambda\propto\|\varepsilon\|_2$ this yields an overall error of order $\|\varepsilon\|_2$; the paper demonstrates numerically on masked Fourier measurements with a PCA generative model of handwritten digits that the interpolated estimator performs best at both high and low signal-to-noise ratios, including for out-of-distribution signals.

Load-bearing premise

The proofs assume that the measurement map and the generator are injective and stretch distances by bounded factors, that the combined map stretches distances less than the measurement map alone, and that the generator is well-conditioned; the numerical experiments do not verify any of these conditions.

Editorial extensions

If this is right

With the coupling parameter chosen as $\lambda\propto\|\varepsilon\|_2$, the unified estimator's reconstruction error is bounded by $C\|\varepsilon\|_2$, matching the noise level up to a constant that may depend on the model bias.
For in-distribution data the generative and unified methods behave similarly, while for out-of-distribution data the unified method avoids the bias floor that limits the pure generative method at high signal-to-noise ratio.
From the generative formulation's residual one can bound the model bias $\|G(\tilde z)-f_0\|_2$ above and below in terms of the noise level, which is useful when the true signal contains defects not representable by the model.
When the noise level is not known in advance, the paper suggests letting the coupling parameter vary during iteration based on the fitting term, making the method deployable without oracle noise knowledge.

Reading between the lines

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

Beyond the paper's claims, the same $\lambda^2\|G(z)-f\|_2^2$ coupling could transfer to other inverse problems with generative priors, such as compressed sensing or deblurring, wherever a bias-vs-variance trade-off can be measured.
A testable next step is to estimate $\alpha$, $\beta$, and $\gamma$ numerically for the actual measurement system before choosing $\lambda$; the paper leaves these constants unspecified for its experiments, and verifying $\gamma<\alpha$ would tell practitioners when generative regularization is worth the bias.
Because the noise model is additive Gaussian, the analysis does not cover Poisson shot noise, which dominates in low-photon ptychography; extending the unified estimator to Poisson noise would require a different fidelity term and is a natural follow-up.

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

Two things you should know. The unified objective (3) is a sensible idea: interpolate between the classical and generative formulations with a single λ. The paper is also honestly written and the three lemmas are correct as conditional statements. But the experimental section does not test the method as analyzed, and the theoretical bound in Lemma 3 does not actually establish the claimed 'bridging' behavior. So the central pitch is plausible but not supported by the evidence as it stands.

What is new: the explicit unified objective, the bias-variance style bounds, and the bias-detection inequality. Lemmas 1 and 2 are standard but useful; Lemma 3 is a simple triangle-inequality bound, not a deep result. The paper is short and readable, and the authors clearly mark their results as preliminary.

The soft spots are real. In Section III they map all three methods into the generic solver (4) and set w=1 everywhere. That adds an extra Tikhonov penalty on z to the unified method and to the generative baseline. They also use λ=σ² for (1) and (2) but λ=10σ² for (3). So Figure 3 compares differently regularized objectives, and the apparent advantage of (3) could be due to the extra z-regularization or the larger λ, not to the coupling term in (3). The stress-test note is right about this.

Second, Lemma 3's bound, ||f~-f0|| ≤ λα||G(z0)-f0|| + 2α||ε||, does not behave like a bridging bound. As λ→0 it gives Lemma 1, but as λ→∞ it blows up unless the bias is exactly zero; it never recovers the generative bound of Lemma 2. So the statement 'as λ→∞ we retrieve (2)' is true for the objective, but the error bound doesn't reflect that. The 'bridging' property is asserted, not proven.

Third, the bi-Lipschitz assumptions (γ<α, well-conditioned G) are never verified for the PCA generator and masked Fourier measurements. That leaves the theory disconnected from the experiment. The authors acknowledge the bounds are crude, but they don't address this gap.

The experiment itself is also minimal: one low-dimensional dataset, no error bars, no multiple runs, no code. For a paper whose main claim is empirical ('achieves the best result'), that's thin.

Who is this for? Someone working on generative priors for inverse problems might find the unified formulation worth trying in their own code, but they should not cite the empirical comparison as evidence. The paper deserves a serious referee because the idea is reasonable and the theory is correct in the conditional sense; but it needs a major revision to actually test the proposed method and to either prove or soften the bridging claim. I'd recommend sending it to review with clear requests for a cleaner experiment, verification or discussion of the assumptions, and code release.

Referee Report

4 major / 6 minor

Summary. The paper studies phase retrieval with a generative-model prior. It compares the classical variational formulation (1), the generative latent-variable formulation (2), and a proposed unified formulation (3) that couples a data term on the object with a penalty on the distance to the generative model. Under bi-Lipschitz assumptions on the measurement map A and the generator G, the authors prove deterministic error bounds for the three formulations (Lemmas 1-3), discuss bias detection, and report a low-dimensional numerical experiment on masked Fourier measurements with a PCA-based generator for MNIST-like data. The central claims are that the unified method bridges (1) and (2) and that it performs best across noise levels on both in-distribution and out-of-distribution data.

Significance. If the conditional bounds were applicable, Lemma 2 would provide a clean bias-variance decomposition for generative regularization, and Lemma 3 would offer a simple error bound for a hybrid objective. The proofs of Lemmas 1-3 are self-contained, and the derivations do not assume the target result, which is a strength. However, the significance is currently limited by three load-bearing gaps: the bi-Lipschitz/injectivity assumptions are not verified and in fact fail for the phase retrieval map as stated; the numerical implementation does not match the analyzed objective; and the error bounds do not actually demonstrate that the unified method inherits the favorable noise behavior of the generative formulation. The experimental evidence is a single low-dimensional run without error bars or code, so the empirical claim is not yet established. The paper is a reasonable starting point, but the central claims require substantial revision.

major comments (4)

[Section II, opening assumptions and Section III-A]
[Section III, implementation of (4)]
[Section III, Figure 3 and surrounding text]
[Section II-C, Lemma 3 and Remark 3]

minor comments (6)

[Section II-C, proof of Lemma 3]
[Section III-A]
[Section II]
[Section II-B]
[Section III, Figure 3]
[Lemma 2 proof]

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the theoretical bounds are self-contained inequalities from stated assumptions, and the single self-citation is motivational rather than load-bearing.

full rationale

The paper's derivation chain is self-contained. Lemma 1 follows from the optimality of the least-squares estimate and the bi-Lipschitz property of A; Lemma 2 follows from optimality, triangle inequalities, and the bi-Lipschitz properties of A, G, and A∘G; Lemma 3 follows from optimality of the unified objective (3) and the inequality sqrt(a^2+b^2) ≤ |a|+|b|, with the bound explicitly depending on the bias term ||G(z0)-f0|| and the noise level ||ε||. None of these proofs assumes the target reconstruction error bound or the claimed superiority of the unified method. The reduction of (1), (2), and (3) to the generic problem (4) is an algebraic reparameterization, not a circular definition. The only self-citation, reference [24], is used to motivate the observation that classical and generative formulations dominate at different signal-to-noise ratios; it is not cited as a proof ingredient or as a uniqueness theorem that forces the paper's choice. The numerical section fits a PCA generative model to MNIST data and manually selects λ and w, which is parameter tuning in the evaluation rather than a fitted quantity being relabeled as a prediction; whether that experimental comparison is well controlled is a validity concern, not circularity. The paper itself notes that its error bounds are crude, which is an acknowledged limitation rather than a circular step. Therefore no circular step can be exhibited from the text.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The theoretical results are conditional on the stated bi-Lipschitz and conditioning assumptions; the numerical results depend on hand-chosen hyperparameters such as λ, k, and ℓ. No new physical or mathematical entities are introduced.

free parameters (4)

λ for combined method (3) = λ = 10·σ²
Set to ten times the noise variance, while both baselines use λ = σ²; this unequal tuning can explain part of the reported improvement in Figure 3.
λ for baseline methods (1) and (2) = λ = σ²
Tikhonov weights chosen for stability; the different value used for (3) confounds the comparison.
latent dimension k of the generative model = k = 30
PCA truncation for the MNIST generative model; it controls the expressiveness and bias of the prior, and the paper does not sweep it.
number of masks ℓ = ℓ = 100
Measurement budget chosen by hand for the experiment; no ablation is reported.

assumptions (5)

domain assumption The measurement map A and the generative model G are injective and bi-Lipschitz with constants α and β.
Section II opening: 'Let us assume that both the measurement map A and the generative model G are injective and bi-Lipschitz...' This underpins all three lemmas.
domain assumption The composition A∘G has a more favorable bi-Lipschitz constant than A, i.e., γ < α.
Section II: 'We furthermore assume that the bi-Lipschitz constant γ of A∘G is more favorable than that of A, that is, γ < α.' Without this, the generative model does not improve conditioning.
domain assumption The generative model is well-conditioned, 0 < β−1 ≪ 1.
Section II: 'G is well-conditioned, in the sense that 0 < β−1 ≪ 1.' This controls variance amplification in Lemma 2 but is not verified for the PCA model.
domain assumption Only additive Gaussian noise is considered, not Poisson shot noise.
Introduction: 'we only consider the additive noise model, which is typical for detector noise and does not cover Poisson shot noise.'
domain assumption Out-of-distribution ground truths have the form f0 = G(z0) + η.
Section II-B introduces this defect model for detecting bias; the bias-detection argument depends on it.

how reviews work

0 comments

Cite this review

Pith. "Pith review of PtyGenography: using generative models for regularization of the phase retrieval problem." pith.science (2026). https://pith.science/paper/MD7TYNVM

@misc{pith2026250201338,
  author       = {Pith},
  title        = {Pith review of: PtyGenography: using generative models for regularization of the phase retrieval problem},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MD7TYNVM}},
  note         = {Machine review of arXiv:2502.01338}
}

read the original abstract

In phase retrieval and similar inverse problems, the stability of solutions across different noise levels is crucial for applications. One approach to promote it is using signal priors in a form of a generative model as a regularization, at the expense of introducing a bias in the reconstruction. In this paper, we explore and compare the reconstruction properties of classical and generative inverse problem formulations. We propose a new unified reconstruction approach that mitigates overfitting to the generative model for varying noise levels.

Figures

Figures reproduced from arXiv: 2502.01338 by the authors.

**Figure 2.** Samples generated by the generative model ( [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Relative reconstruction error for the three methods [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

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

Phasebook: A Survey of Selected Open Problems in Phase Retrieval
cs.IT 2025-05 conditional novelty 3.0 of 10

A workshop-based survey of open problems in phase retrieval, with a section proposing a unified framework that combines generative priors with conventional data fidelity.

Reference graph

Works this paper leans on

26 extracted references · 24 canonical work pages · cited by 1 Pith paper

[1]

Investigation of composite materials using slm-based phase retrieval

Mostafa Agour, Claas Falldorf, and Ralf B Bergmann. Investigation of composite materials using slm-based phase retrieval. Optics Letters, 38(13):2203–2205, 2013

work page 2013
[2]

On signal reconstruc- tion without phase

Radu Balan, Pete Casazza, and Dan Edidin. On signal reconstruc- tion without phase. Applied and Computational Harmonic Analysis , 20(3):345–356, 2006

work page 2006
[3]

On lipschitz analysis and lipschitz synthesis for the phase retrieval problem

Radu Balan and Dongmian Zou. On lipschitz analysis and lipschitz synthesis for the phase retrieval problem. Linear Algebra and its Applications, 496:152–181, 2016

work page 2016
[4]

Phase retrieval with semi-algebraic and ReLU neural network priors

Tamir Bendory, Nadav Dym, Dan Edidin, and Arun Suresh. Phase retrieval with semi-algebraic and relu neural network priors. arXiv preprint arXiv:2311.08833, 2023

work page Pith review arXiv 2023
[5]

Diffractive imaging for periodic samples: retrieving one-dimensional concentration profiles across microfluidic channels

Oliver Bunk, Ana Diaz, Franz Pfeiffer, Christian David, Bernd Schmitt, Dillip K Satapathy, and J Friso Van Der Veen. Diffractive imaging for periodic samples: retrieving one-dimensional concentration profiles across microfluidic channels. Acta Crystallographica Section A: Foun- dations of Crystallography , 63(4):306–314, 2007

work page 2007
[6]

The mnist database of handwritten digit images for machine learning research

Li Deng. The mnist database of handwritten digit images for machine learning research. IEEE Signal Processing Magazine , 29(6):141–142, 2012

work page 2012
[7]

Learning to synthesize: robust phase retrieval at low photon counts

Mo Deng, Shuai Li, Alexandre Goy, Iksung Kang, and George Bar- bastathis. Learning to synthesize: robust phase retrieval at low photon counts. Light: Science & Applications , 9(1):36, 2020

work page 2020
[8]

Phase retrieval: From computational imaging to machine learning: A tutorial

Jonathan Dong, Lorenzo Valzania, Antoine Maillard, Thanh An Pham, Sylvain Gigan, and Michael Unser. Phase retrieval: From computational imaging to machine learning: A tutorial. IEEE Signal Processing Magazine, 40:45–57, 2023

work page 2023

Show all 26 references

[9]

Sparse phase retrieval from short-time fourier measure- ments

Yonina C Eldar, Pavel Sidorenko, Dustin G Mixon, Shaby Barel, and Oren Cohen. Sparse phase retrieval from short-time fourier measure- ments. IEEE Signal Processing Letters , 22(5):638–642, 2014

2014
[10]

Phase retrieval: Uniqueness and stability

Philipp Grohs, Sarah Koppensteiner, and Martin Rathmair. Phase retrieval: Uniqueness and stability. SIAM Review, 62:301–350, 2020

2020
[11]

Phase retrieval under a generative prior

Paul Hand, Oscar Leong, and Vlad V oroninski. Phase retrieval under a generative prior. Advances in Neural Information Processing Systems , 31, 2018

2018
[12]

Quantum tomography under prior information

Teiko Heinosaari, Luca Mazzarella, and Michael M Wolf. Quantum tomography under prior information. Communications in Mathematical Physics, 318(2):355–374, 2013

2013
[13]

Lower lipschitz bounds for phase retrieval from locally supported measurements.Applied and Computational Harmonic Analysis , 47(2):526–538, 2019

Mark A Iwen, Sami Merhi, and Michael Perlmutter. Lower lipschitz bounds for phase retrieval from locally supported measurements.Applied and Computational Harmonic Analysis , 47(2):526–538, 2019

2019
[14]

Sparse phase retrieval: Uniqueness guarantees and recovery algorithms

Kishore Jaganathan, Samet Oymak, and Babak Hassibi. Sparse phase retrieval: Uniqueness guarantees and recovery algorithms. IEEE Trans- actions on Signal Processing , 65(9):2402–2410, 2017

2017
[15]

Development of hybrid optical sensor based on deep learning to detect and classify the micro-size defects in printed circuit board

Gulhan Ustabas Kaya. Development of hybrid optical sensor based on deep learning to detect and classify the micro-size defects in printed circuit board. Measurement, 206:112247, 2023

2023
[16]

Characterization of surface defects using a phase retrieval technique in a high-power laser system

Lucien Lehmann, Stéphane Bouillet, Christophe Leymarie, Christel Ameil-Schuh, Melusine Benoit, and Claude Rouyer. Characterization of surface defects using a phase retrieval technique in a high-power laser system. Applied Optics, 61(6):1545–1551, 2022

2022
[17]

On the limited memory bfgs method for large scale optimization

Dong C Liu and Jorge Nocedal. On the limited memory bfgs method for large scale optimization. Mathematical programming , 45(1):503– 528, 1989

1989
[18]

Phase retrieval in crystallography and optics

Rick P Millane. Phase retrieval in crystallography and optics. JOSA A, 7(3):394–411, 1990

1990
[19]

Actinic imaging and evaluation of phase structures on extreme ultraviolet lithography masks

Iacopo Mochi, Kenneth A Goldberg, and Sungmin Huh. Actinic imaging and evaluation of phase structures on extreme ultraviolet lithography masks. Journal of Vacuum Science & Technology B , 28(6):C6E11– C6E16, 2010

2010
[20]

Fundamentals of speech recognition

Lawrence R Rabiner and Biing-Hwang Juang. Fundamentals of speech recognition. Tsinghua University Press, 1999

1999
[21]

Phase recovery and holographic image reconstruction using deep learning in neural networks

Yair Rivenson, Yibo Zhang, Harun Günaydın, Da Teng, and Aydogan Ozcan. Phase recovery and holographic image reconstruction using deep learning in neural networks. Light: Science & Applications, 7(2):17141– 17141, 2018

2018
[22]

Ptychography and related diffractive imaging methods

John M Rodenburg. Ptychography and related diffractive imaging methods. Advances in imaging and electron physics , 150:87–184, 2008

2008
[23]

Sparse pursuit and dictionary learning for blind source separation in polyphonic music recordings

Sören Schulze and Emily J King. Sparse pursuit and dictionary learning for blind source separation in polyphonic music recordings. EURASIP Journal on Audio, Speech, and Music Processing , 2021(1):1–25, 2021

2021
[24]

Jacob Seifert, Yifeng Shao, and Allard P. Mosk. Noise-robust latent vector reconstruction in ptychography using deep generative models. Optics Express, 32:1020, 1 2024

2024
[25]

Lensless computational imaging through deep learning

Ayan Sinha, Justin Lee, Shuai Li, and George Barbastathis. Lensless computational imaging through deep learning. Optica, 4(9):1117–1125, 2017

2017
[26]

Some tendencies in the tikhonov regularization of ill-posed problems

VV Vasin. Some tendencies in the tikhonov regularization of ill-posed problems. Journal of Inverse & Ill-Posed Problems , 14(8), 2006

2006

Pith tools

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

[1] [1]

Investigation of composite materials using slm-based phase retrieval

Mostafa Agour, Claas Falldorf, and Ralf B Bergmann. Investigation of composite materials using slm-based phase retrieval. Optics Letters, 38(13):2203–2205, 2013

work page 2013

[2] [2]

On signal reconstruc- tion without phase

Radu Balan, Pete Casazza, and Dan Edidin. On signal reconstruc- tion without phase. Applied and Computational Harmonic Analysis , 20(3):345–356, 2006

work page 2006

[3] [3]

On lipschitz analysis and lipschitz synthesis for the phase retrieval problem

Radu Balan and Dongmian Zou. On lipschitz analysis and lipschitz synthesis for the phase retrieval problem. Linear Algebra and its Applications, 496:152–181, 2016

work page 2016

[4] [4]

Phase retrieval with semi-algebraic and ReLU neural network priors

Tamir Bendory, Nadav Dym, Dan Edidin, and Arun Suresh. Phase retrieval with semi-algebraic and relu neural network priors. arXiv preprint arXiv:2311.08833, 2023

work page Pith review arXiv 2023

[5] [5]

Diffractive imaging for periodic samples: retrieving one-dimensional concentration profiles across microfluidic channels

Oliver Bunk, Ana Diaz, Franz Pfeiffer, Christian David, Bernd Schmitt, Dillip K Satapathy, and J Friso Van Der Veen. Diffractive imaging for periodic samples: retrieving one-dimensional concentration profiles across microfluidic channels. Acta Crystallographica Section A: Foun- dations of Crystallography , 63(4):306–314, 2007

work page 2007

[6] [6]

The mnist database of handwritten digit images for machine learning research

Li Deng. The mnist database of handwritten digit images for machine learning research. IEEE Signal Processing Magazine , 29(6):141–142, 2012

work page 2012

[7] [7]

Learning to synthesize: robust phase retrieval at low photon counts

Mo Deng, Shuai Li, Alexandre Goy, Iksung Kang, and George Bar- bastathis. Learning to synthesize: robust phase retrieval at low photon counts. Light: Science & Applications , 9(1):36, 2020

work page 2020

[8] [8]

Phase retrieval: From computational imaging to machine learning: A tutorial

Jonathan Dong, Lorenzo Valzania, Antoine Maillard, Thanh An Pham, Sylvain Gigan, and Michael Unser. Phase retrieval: From computational imaging to machine learning: A tutorial. IEEE Signal Processing Magazine, 40:45–57, 2023

work page 2023