Pith. sign in

REVIEW 4 major objections 6 minor 33 references

Orientation Determination of Cryo-EM Images Using Block Stochastic Riemannian Subgradient Methods

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read A block stochastic Riemannian subgradient method solves the LUD cryo-EM orientation problem with accuracy matching state-of-the-art solvers while running roughly 20 times faster on average.

desk verdict A genuinely useful empirical paper with a real algorithmic synthesis, but the main convergence theorem analyzes a different algorithm than the one benchmarked. read the letter →

arxiv 2411.14021 v1 pith:SGMZSACR submitted 2024-11-21 math.OC

Wanyu Zhang , Ruili Gou , Huikang Liu , Zhiguo Wang , Yinyu Ye This is my paper · ORCID
classification math.OC MSC 90C2665K0590C15
keywords cryo-EMorientationdeterminationcommon-linemethodleastunsquareddeviation(LUD)RiemanniansubgradientblockstochasticoptimizationStiefelmanifoldlowsignal-to-noiseratio
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 show that the least unsquared deviation (LUD) formulation of cryo-EM orientation determination — a nonconvex, nonsmooth synchronization problem over rotations — can be solved directly by a Riemannian subgradient method, without semidefinite relaxation. The block stochastic variant updates only a small random subset of rotations using a small random subset of common-line measurements, and the paper claims this preserves accuracy while giving an average 20-fold speedup over state-of-the-art methods. It would matter because real cryo-EM datasets contain hundreds of thousands to millions of images, and existing solvers are too slow at that scale. The paper also claims all proposed variants reach an $\varepsilon$-approximate stationary point in $O(\varepsilon^{-4})$ iterations, and that a spectral-norm-constrained version keeps accuracy at low signal-to-noise ratios where unconstrained methods fail.

What carries the argument

The load-bearing object is the block stochastic Riemannian subgradient oracle on the Stiefel manifold $\mathrm{St}(3K, 3)$ treated as a product of $K$ copies of $SO(3)$, where each oracle call draws random subsets $D$ and $S$ of rotations and common lines and returns a partial Euclidean subgradient of the form $\sum_{j \in S} (R_i c_{ij} - R_j c_{ji}) c_{ij}^\top / \|R_i c_{ij} - R_j c_{ji}\|_2$. Two tools do the theoretical work: the weak convexity inequality for functions on the Stiefel manifold, imported from Li et al. [2021], which converts one subgradient step into guaranteed descent of the Moreau envelope, and the Moreau-envelope stationarity measure $\Theta(X) = \lambda^{-1} \|P_\lambda f(X) - X\|_F$, whose smallness is what the $O(\varepsilon^{-4})$ bound controls. The QR-based retraction keeps every update feasible, and it is also where the alternating projection for the spectral norm constraint plugs in.

What would settle it

Re-run ReSync-BSGD on one fixed synthetic dataset twice — once with random reshuffling of blocks per pass, once with independent uniform block partitions — and compare the stationarity measure $\Theta(X_t)$ against the claimed bound $\mathbb{E}[\Theta^2(X_t)] \leq c_1(f_\lambda(X_0) - \min f_\lambda) + c_2/\sqrt{T+1}$; if the reshuffled run stops tracking the bound as $T$ grows, the $O(\varepsilon^{-4})$ result covers an oracle the algorithm does not use. A second check is in the paper's own Table 1: at detection rate $p = 0.05$ every method lands at MSE $\approx 0.3$–$0.45$, so the 'comparable accuracy' claim should be re-tested in exactly that low-$p$ regime before it is generalized.

Watch

Extended reading notes

Core claim

The central claim is that minimizing $f(R) = \sum_{i,j} \|R_i c_{ij} - R_j c_{ji}\|_2$ over $R_1, \dots, R_K \in SO(3)$ — the LUD common-line objective — can be done by repeated Riemannian subgradient steps: compute a Euclidean subgradient, project it onto the tangent space at $R_i$ via $P_{T_{R_i}}(B) = R_i(R_i^\top B - B^\top R_i)/2$, and map back onto the manifold with a QR-based retraction. The paper claims that sampling small blocks of rotations and small blocks of common lines each iteration (ReSync-BSGD, with filter ratio $\rho$) does not sacrifice the $O(\varepsilon^{-4})$ iteration complexity of the full method, and that in experiments it matches the MSE of semidefinite relaxation and IRLS/PGD solvers while running 5 to 30 times faster. The same machinery, with an added spectral norm constraint enforced by alternating projection, is claimed to prevent viewing directions from clustering at SNR as low as 1/64, and with class-averaging preprocessing even 1/128 becomes tractable.

Load-bearing premise

The $O(\varepsilon^{-4})$ guarantee holds only if each iteration draws its random partition of the rotations independently, but the implemented algorithm reshuffles the block order within each pass, so the proven bound does not formally cover the variant that was benchmarked.

Editorial extensions

If this is right

  • The LUD formulation, previously solved by expensive semidefinite relaxation, becomes solvable by a direct first-order method, so robustness to misdetected common lines is no longer purchased at the price of scalability.
  • All four variants — full ReSync, ReSync-SGD, ReSync-BCD, and ReSync-BSGD — inherit the $O(\varepsilon^{-4})$ complexity bound, since the first three are special cases of the block stochastic method.
  • Because BSGD updates independent blocks in each iteration, the method parallelizes across the $K$ images, which the paper cites as a reason it should scale to the large datasets of real cryo-EM pipelines.
  • The spectral-norm-constrained variant (ReSync-norm) keeps orientation accuracy at SNR 1/64 where unconstrained benchmarks diverge, and combined with class averaging it produces usable results at SNR 1/128, a regime the paper says has not previously been documented.

Reading between the lines

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

  • The independence gap is worth probing: the proof assumes i.i.d. uniform block partitions, while the tested algorithm reshuffles once per pass; the empirical speedup may survive, but proving the $O(\varepsilon^{-4})$ rate for the reshuffled variant is an open task that the paper does not address.
  • The block structure suggests embarrassingly parallel GPU-scale implementation; the reported 20-fold speedup on a 16-core CPU is likely a lower bound on what the method could achieve on larger parallel hardware.
  • The accuracy comparisons concentrate on regimes where the solver matters ($p \geq 0.1$, SNR $\geq 1/64$); at the hardest settings the tables show all methods failing together, so the practical win at extreme noise appears to come from the denoising preprocessing rather than from the optimization method itself.
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.

Referee Report

4 major / 6 minor

Summary. The paper addresses the cryo-EM orientation determination problem using the least unsquared deviation (LUD) formulation. It proposes a Riemannian subgradient method and three stochastic variants (ReSync-SGD, ReSync-BCD, ReSync-BSGD) that update only a subset of rotations and/or use a subset of common lines. For these methods the paper claims an O(ε^-4) iteration complexity to reach an ε-nearly stationary point measured via the Moreau envelope. The same framework is extended to a low-SNR setting by adding a spectral norm constraint, yielding a ReSync-norm algorithm. Numerical experiments on synthetic and simulated real projection data report accuracy comparable to existing methods with a 5- to 30-fold speedup for ReSync-BSGD.

Significance. If the convergence result held for the implemented algorithms, this would be a useful contribution: a first iteration-complexity bound for block stochastic Riemannian subgradient methods in this application, plus a reproducible practical speedup. The paper's strengths are its open-source code, its experiments across detection rates and SNRs down to 1/128, its explicit comparison with SDR-ADMM, and a self-contained proof strategy using the Moreau-envelope stationarity measure. However, the central theoretical claim is currently proven for an oracle that does not match Algorithm 2, so the formal guarantee does not cover the method that is benchmarked. The empirical claims may be valid, but the theoretical gap is load-bearing and needs to be closed in revision.

major comments (4)
  1. [Appendix A.1 and Proposition 1 vs. Section 3.3 and Algorithm 2] The convergence analysis is carried out for an oracle that differs structurally from the implemented algorithm. Appendix A.1 defines m = ρK blocks, each of size 1/ρ, and the approximate subgradient g_i(X_i, ζ) sums f_{i,j} over the block containing i, yielding the bound ||∇g_i|| ≤ B = 1/ρ used in Proposition 1 and in the constant c_2 in (23). Algorithm 2, by contrast, updates a subset D of size ρ_1 K using a data subset S of size ρ_2 K; for ReSync-BSGD, |D| = |S| = ρK and the update for i ∈ D sums over j ∈ S, so with unit-norm common-line vectors the subgradient norm is at most ρK (e.g., 300 when K = 3000 and ρ = 0.1), not 1/ρ = 10. Consequently, Proposition 1's descent inequality and the resulting O(ε^-4) bound in Theorem 1 are not established for the algorithm actually tested in Tables 1-4.
  2. [Appendix A.1, property (a), and Algorithm 2] The proof assumes an oracle that generates independent and identically distributed partitions ζ_t ~ U(Ω), but Algorithm 2 and the surrounding text in Section 3.3 describe random reshuffling of {1, ..., K} before each iteration. Random reshuffling induces dependence between successive partitions, so the telescoping sum over ζ_0, ..., ζ_{T-1} in the proof of Theorem 1 is not justified for the algorithm as written. This is a second, independent gap: even after correcting the block-size mismatch, the proof would need a random-reshuffling analysis or the algorithm would need to be changed to sample partitions i.i.d.
  3. [Section 4 and Theorem 2] The proof of Theorem 1 relies on the τ-weak convexity inequality stated as Theorem 2, but the paper never verifies that the LUD objective (3) satisfies the required τ-weak convexity condition on Stiefel(3K, 3) with an explicit constant τ. The sentence 'We can verify that the point P_λ f(X) satisfies...' concerns first-order optimality, not weak convexity. Since the entire descent argument depends on this assumption, the authors should either prove the needed weak-convexity property for f or cite a theorem that applies directly to this objective.
  4. [Section 3.4 and problem (11)] The surrogate constraint is stated inconsistently and without justification. Since eR ∈ R^{2K×2}, the matrix eR^T eR is 2×2, so writing 'eR^T eR = αK I_3' in problem (11) is dimensionally incoherent. The text also alternates between the spectral norm constraint ||G||_2 ≤ αK on G = eR eR^T and the surrogate eR^T eR = αK I_3; the equivalence between these constraints is asserted but never proved. Because ReSync-norm is presented as a contribution for low-SNR scenarios, this formulation needs to be corrected and justified.
minor comments (6)
  1. [Algorithm 1, line 7] The displayed update in line 7 contains a garbled symbol 'µk e∇f (Rt_i)'; it should be µ_t \tilde∇ f(R_i^t).
  2. [Abstract and Section 5.1] The phrase 'an averaging 20-fold speedup' should read 'an average 20-fold speedup'.
  3. [Tables 1-3] The method called LUD-IRLS in Tables 1 and 2 is called LUD-IRLS-PGD in Table 3; please use a consistent name.
  4. [Table 2] For K = 5000 and SNR = 1/32, the Eig row reports MSE 4.19E+00, which is markedly different from the neighboring Eig rows (3.08E-02 at SNR=1/16 and 4.86E+00 at SNR=1/64); please verify this entry and comment on its cause.
  5. [Appendix A.1, after the long inequality] The text says 'The third equation uses Lemma 1...', but the displayed chain contains inequalities, not numbered equations; this should be reworded as 'third inequality'.
  6. [Problem (11)] The constraint '1 ≤ i ≤ K' under the surrogate eR^T eR = αK I_3 is unclear because eR^T eR is a global quantity, not per-image; the notation should be clarified.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: convergence analysis is self-contained and empirical claims are benchmarked externally.

full rationale

The paper's central derivation—the O(epsilon^-4) iteration complexity for the block stochastic Riemannian subgradient method—is an independent mathematical argument built on the weak-convexity framework of Li et al. (2021) and Davis and Drusvyatskiy (2019). It does not define its stationarity measure or convergence claim in terms of fitted parameters, and no experimental quantity is renamed as a prediction. The ReSync framework is inherited from the authors' prior work, but that citation is not load-bearing: Algorithm 2 is specified directly in the paper, and the convergence proof does not rely on Liu et al. (2023) for its conclusion. The reported 5-to-30-fold speedups are comparisons against standard external baselines (eigenvector relaxation, PGD, IRLS) in Tables 1 and 2, so the empirical claims are externally falsifiable. A possible gap between the analyzed i.i.d. block oracle in Appendix A.1, which uses blocks of size 1/rho and the bound B=1/rho, and the practically implemented random-reshuffling subsets D and S of size rho*K is a correctness or rigor concern, not a circularity: the proof may fail to cover Algorithm 2 as written, but it does not assume the conclusion it purports to prove. The limitations section candidly notes that the convergence-rate superiority over SGD and BCD is not established, which is a scope caveat rather than a circular step. No fitted input is relabeled as a prediction, no uniqueness theorem is imported from the authors' own prior work, and no known result is merely renamed. The derivation chain is therefore not circular.

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

The paper introduces no new physical entities or conserved quantities. Its contributions rest on standard weakly convex analysis, retraction properties, and domain assumptions about common-line geometry. The main entries are algorithmic hyperparameters (rho, alpha, step size, stopping tolerances) and the i.i.d. oracle assumption that is stronger than the implemented random reshuffling.

free parameters (4)
  • filter ratio rho = 0.1 in experiments
    Controls the fraction of rotations and common-line data used per iteration in ReSync-BSGD, SGD, and BCD variants. Chosen by hand; affects the accuracy-speed tradeoff shown in Figure 3.
  • spectral norm hyperparameter alpha = 2/3
    Used in the norm-constrained reformulation (11). Set to 2/3 based on prior knowledge that viewing directions are uniformly distributed, as stated in Section 5.2.1.
  • step size schedule mu_t = 1/sqrt(T+1)
    Constant step size schedule that depends on the total iteration count T. Standard for stochastic subgradient methods but requires knowing T in advance.
  • stopping precision thresholds = e.g., 5e-7 for p=0.5, 5e-6 for p=0.3
    Target MSE/precision used to terminate each method in Table 1; these differ across detection rates and are chosen to allow fair comparison within the maximum iteration count.
assumptions (5)
  • standard math The LUD objective (3) is weakly convex on the Stiefel manifold St(3K, 3), allowing use of the weak convexity inequality from Theorem 2 of Li et al. 2021.
    Invoked in Appendix A before Lemma 1 and in the proof of Theorem 2. The authors do not verify the weak convexity constant tau or the Lipschitz constant L for the specific common-line terms; they rely on the cited theorem.
  • standard math The QR-decomposition based retraction satisfies the second-order boundedness property with a constant b, and M = max{B^2 b_i} is finite.
    Used in Lemma 1 and in the bound of Proposition 1. Standard for compact Stiefel manifolds but stated without proof in this paper.
  • domain assumption The stochastic oracle generates independent and identically distributed samples uniformly over the space of block partitions.
    Property (a) in Section A.1. This is the load-bearing assumption for the telescoping sum in Theorem 1; it is not satisfied by the random reshuffling implementation in Algorithm 2.
  • domain assumption Detected common-line vectors c_ij and c_ji are unit vectors and satisfy R_i c_ij = R_j c_ji for ground-truth rotations R_i, R_j.
    This is the basic common-line consistency relation (1) used to define the objective. It is an idealization; in real data the detected lines are noisy and may be outliers, which is why the LUD formulation is used.
  • ad hoc to paper The surrogate constraint eR^T eR = alpha K I_3 is a valid replacement for the spectral norm constraint ||G||_2 <= alpha K on the Gram matrix G = eR eR^T.
    Introduced in Section 3.4, equation (11). The authors state this is a surrogate; no derivation shows that solving the surrogates gives the same solutions as the original spectral-norm-constrained problem.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Orientation Determination of Cryo-EM Images Using Block Stochastic Riemannian Subgradient Methods." pith.science (2026). https://pith.science/paper/SGMZSACR

@misc{pith2026241114021,
  author       = {Pith},
  title        = {Pith review of: Orientation Determination of Cryo-EM Images Using Block Stochastic Riemannian Subgradient Methods},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SGMZSACR}},
  note         = {Machine review of arXiv:2411.14021}
}
read the original abstract

The determination of molecular orientations is crucial for the three-dimensional reconstruction of Cryo-EM images. Traditionally addressed using the common-line method, this challenge is reformulated as a self-consistency error minimization problem constrained to rotation groups. In this paper, we consider the least-squared deviation (LUD) formulation and employ a Riemannian subgradient method to effectively solve the orientation determination problem. To enhance computational efficiency, a block stochastic version of the method is proposed, and its convergence properties are rigorously established. Extensive numerical evaluations reveal that our method not only achieves accuracy comparable to that of state-of-the-art methods but also delivers an average 20-fold speedup. Additionally, we implement a modified formulation and algorithm specifically designed to address scenarios characterized by very low SNR.

Figures

Figures reproduced from arXiv: 2411.14021 by the authors.

Figure 1
Figure 1. 2D visualization of viewing directions by ground-truth and ReSync algorithms ( [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Comparison with other orientation determination algorithms under synthetic data ( [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Convergence of ReSync-BSGD under synthetic data ( [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Projection images of 50-S ribosomal subunit under different SNR levels. [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

33 extracted references · 29 canonical work pages

  1. [1]

    Optimization algorithms on matrix manifolds

    P-A Absil, Robert Mahony, and Rodolphe Sepulchre. Optimization algorithms on matrix manifolds. Princeton University Press, 2008

    2008

  2. [2]

    Backpropagation and stochastic gradient descent method

    Shun-ichi Amari. Backpropagation and stochastic gradient descent method. Neurocomputing, 5 0 (4-5): 0 185--196, 1993

    1993

  3. [3]

    Non-unique games over compact groups and orientation estimation in cryo-em

    Afonso S Bandeira, Yutong Chen, Roy R Lederman, and Amit Singer. Non-unique games over compact groups and orientation estimation in cryo-em. Inverse Problems, 36 0 (6): 0 064002, 2020

    work page 2020

  4. [4]

    Single-particle cryo-electron microscopy: Mathematical theory, computational challenges, and opportunities

    Tamir Bendory, Alberto Bartesaghi, and Amit Singer. Single-particle cryo-electron microscopy: Mathematical theory, computational challenges, and opportunities. IEEE signal processing magazine, 37 0 (2): 0 58--76, 2020

    work page 2020

  5. [5]

    Proximal gradient method for nonsmooth optimization over the stiefel manifold

    Shixiang Chen, Shiqian Ma, Anthony Man-Cho So, and Tong Zhang. Proximal gradient method for nonsmooth optimization over the stiefel manifold. SIAM Journal on Optimization, 30 0 (1): 0 210--239, 2020

    2020

  6. [6]

    Three dimensional reconstructions of spherical viruses by fourier synthesis from electron micrographs

    RAea Crowther, Linda A Amos, JT Finch, DJ De Rosier, and A Klug. Three dimensional reconstructions of spherical viruses by fourier synthesis from electron micrographs. Nature, 226 0 (5244): 0 421--425, 1970

    work page 1970

  7. [7]

    Stochastic model-based minimization of weakly convex functions

    Damek Davis and Dmitriy Drusvyatskiy. Stochastic model-based minimization of weakly convex functions. SIAM Journal on Optimization, 29 0 (1): 0 207--239, 2019

    2019

  8. [8]

    Efficiency of minimizing compositions of convex functions and smooth maps

    Dmitriy Drusvyatskiy and Courtney Paquette. Efficiency of minimizing compositions of convex functions and smooth maps. Mathematical Programming, 178: 0 503--558, 2019

    work page 2019

Show all 33 references
  1. [9]

    Fast fourier transforms for nonequispaced data

    Alok Dutt and Vladimir Rokhlin. Fast fourier transforms for nonequispaced data. SIAM Journal on Scientific computing, 14 0 (6): 0 1368--1393, 1993

    1993

  2. [10]

    Nonuniform fast fourier transforms using min-max interpolation

    Jeffrey A Fessler and Bradley P Sutton. Nonuniform fast fourier transforms using min-max interpolation. IEEE transactions on signal processing, 51 0 (2): 0 560--574, 2003

    2003

  3. [11]

    Cryo-electron microscopy as an investigative tool: the ribosome as an example

    Joachim Frank. Cryo-electron microscopy as an investigative tool: the ribosome as an example. Bioessays, 23 0 (8): 0 725--732, 2001

    2001

  4. [12]

    Three-dimensional electron microscopy of macromolecular assemblies: visualization of biological molecules in their native state

    Joachim Frank. Three-dimensional electron microscopy of macromolecular assemblies: visualization of biological molecules in their native state. Oxford university press, 2006

    2006

  5. [13]

    Mathematics for computer tomography

    Bertil Gustafsson. Mathematics for computer tomography. Physica Scripta, 1996 0 (T61): 0 38, 1996

    1996

  6. [14]

    Riemannian proximal gradient methods

    Wen Huang and Ke Wei. Riemannian proximal gradient methods. Mathematical Programming, 194 0 (1): 0 371--413, 2022

    2022

  7. [15]

    Weakly convex optimization over stiefel manifold using riemannian subgradient-type methods

    Xiao Li, Shixiang Chen, Zengde Deng, Qing Qu, Zhihui Zhu, and Anthony Man-Cho So. Weakly convex optimization over stiefel manifold using riemannian subgradient-type methods. SIAM Journal on Optimization, 31 0 (3): 0 1605--1634, 2021

    2021

  8. [16]

    Convergence of random reshuffling under the kurdyka-- ojasiewicz inequality

    Xiao Li, Andre Milzarek, and Junwen Qiu. Convergence of random reshuffling under the kurdyka-- ojasiewicz inequality. SIAM Journal on Optimization, 33 0 (2): 0 1092--1120, 2023

    2023

  9. [17]

    Resync: Riemannian subgradient-based robust rotation synchronization

    Huikang Liu, Xiao Li, and Anthony Man-Cho So. Resync: Riemannian subgradient-based robust rotation synchronization. arXiv preprint arXiv:2305.15136, 2023

    2023 arXiv

  10. [18]

    Challenges and opportunities in cryo-em single-particle analysis

    Dmitry Lyumkis. Challenges and opportunities in cryo-em single-particle analysis. Journal of Biological Chemistry, 294 0 (13): 0 5181--5197, 2019

    2019

  11. [19]

    Orientation estimation of cryo-em images using projected gradient descent method

    Huan Pan, Jian Lu, You-Wei Wen, Chen Xu, and Tieyong Zeng. Orientation estimation of cryo-em images using projected gradient descent method. Inverse Problems, 39 0 (4): 0 045002, 2023

    2023

  12. [20]

    A new 3-d reconstruction scheme applied to the 50s ribosomal subunit of e

    M Radermacher, T Wagenknecht, A Verschoor, and J Frank. A new 3-d reconstruction scheme applied to the 50s ribosomal subunit of e. coli. Journal of microscopy, 141 0 (1): 0 RP1--RP2, 1986

    1986

  13. [21]

    Peter B Rosenthal. From high symmetry to high resolution in biological electron microscopy: a commentary on crowther (1971)‘procedures for three-dimensional reconstruction of spherical viruses by fourier synthesis from electron micrographs’. Philosophical Transactions of the R...

    1971

  14. [22]

    How good is sgd with random shuffling? In Conference on Learning Theory, pages 3250--3284

    Itay Safran and Ohad Shamir. How good is sgd with random shuffling? In Conference on Learning Theory, pages 3250--3284. PMLR, 2020

    2020

  15. [23]

    Viewing direction estimation in cryo-em using synchronization

    Yoel Shkolnisky and Amit Singer. Viewing direction estimation in cryo-em using synchronization. SIAM journal on imaging sciences, 5 0 (3): 0 1088--1110, 2012

    2012

  16. [24]

    Three-dimensional structure determination from common lines in cryo-em by eigenvectors and semidefinite programming

    Amit Singer and Yoel Shkolnisky. Three-dimensional structure determination from common lines in cryo-em by eigenvectors and semidefinite programming. SIAM journal on imaging sciences, 4 0 (2): 0 543--572, 2011

    2011

  17. [25]

    Vector diffusion maps and the connection laplacian

    Amit Singer and H-T Wu. Vector diffusion maps and the connection laplacian. Communications on pure and applied mathematics, 65 0 (8): 0 1067--1144, 2012

    2012

  18. [26]

    Detecting consistent common lines in cryo-em by voting

    Amit Singer, Ronald R Coifman, Fred J Sigworth, David W Chester, and Yoel Shkolnisky. Detecting consistent common lines in cryo-em by voting. Journal of structural biology, 169 0 (3): 0 312--322, 2010

    2010

  19. [27]

    Convergence of a block coordinate descent method for nondifferentiable minimization

    Paul Tseng. Convergence of a block coordinate descent method for nondifferentiable minimization. Journal of optimization theory and applications, 109: 0 475--494, 2001

    2001

  20. [28]

    Angular reconstitution: a posteriori assignment of projection directions for 3d reconstruction

    Marin Van Heel. Angular reconstitution: a posteriori assignment of projection directions for 3d reconstruction. Ultramicroscopy, 21 0 (2): 0 111--123, 1987

    1987

  21. [29]

    Orientation determination of cryo-em images using least unsquared deviations

    Lanhui Wang, Amit Singer, and Zaiwen Wen. Orientation determination of cryo-em images using least unsquared deviations. SIAM journal on imaging sciences, 6 0 (4): 0 2450--2483, 2013

    2013

  22. [30]

    Understanding the invisible hands of sample preparation for cryo-em

    Giulia Weissenberger, Rene JM Henderikx, and Peter J Peters. Understanding the invisible hands of sample preparation for cryo-em. Nature Methods, 18 0 (5): 0 463--471, 2021

    2021

  23. [31]

    Block stochastic gradient iteration for convex and nonconvex optimization

    Yangyang Xu and Wotao Yin. Block stochastic gradient iteration for convex and nonconvex optimization. SIAM Journal on Optimization, 25 0 (3): 0 1686--1716, 2015

    2015

  24. [32]

    Optimality conditions for the nonlinear programming problems on riemannian manifolds

    Wei Hong Yang, Lei-Hong Zhang, and Ruyi Song. Optimality conditions for the nonlinear programming problems on riemannian manifolds. Pacific Journal of Optimization, 10 0 (2): 0 415--434, 2014

    2014

  25. [33]

    Accelerated mini-batch randomized block coordinate descent method

    Tuo Zhao, Mo Yu, Yiming Wang, Raman Arora, and Han Liu. Accelerated mini-batch randomized block coordinate descent method. Advances in neural information processing systems, 27, 2014

    2014

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.