REVIEW 3 major objections 7 minor 36 references
Local Averaging Accurately Distills Manifold Structure From Noisy Data
T0 review · 3 major / 7 minor · reviewed 2026-08-15 · deepseek-v4-flash
Pith's one-line read This paper proves that two rounds of local averaging bring noisy points to within $O(\sigma\sqrt{d})$ of a low-dimensional manifold, even when the noise is comparable to the manifold's reach.
desk verdict Genuinely useful subgaussian phase-transition tool and a credible Stage I analysis, but the advertised Stage II regime is not proven and the paper overstates it. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing object is the grouping probability $h(s)=P(g_2^2+\cdots+g_D^2\le (R^2-s^2)/\sigma^2)$ for i.i.d. standard Gaussians $g_i$, which is the probability that a clean point at distance $s$ from the current landmark is accepted into the ball. Its sharp phase transition at $s_\star^2=R^2-\sigma^2(D-3)$ captures the geometry of acceptance: in high ambient dimension the noise shell makes points at that distance the main contributors. The paper derives two-sided subgaussian bounds for $h$ and its derivative near $s_\star$, convolves them with the one-dimensional Gaussian along the landmark direction to bound the conditional mean of accepted noise, and combines this with a curvature lemma that bounds the signal average's distance to the manifold by $\kappa$ times the average squared intrinsic distance. An injected Gaussian perturbation after round one keeps the landmark at distance at least $\sigma D^{3/8}$ from the manifold, so the second-round analysis stays in the normal-injectivity neighborhood where projection derivatives are controlled.
What would settle it
Run the two-round algorithm on a low-curvature manifold with noise chosen so that $\sigma\sqrt{D}\approx\tau$ but $\sigma\sqrt{D\log D}>1/\kappa$, using the theorem's stated radii and minibatch sizes, and measure $d(q_2,\mathcal{M})$ over many trials; if the distance exceeds $C\sigma\sqrt{d(1+\kappa\,\mathrm{diam}(\mathcal{M})/\log D)}$ with non-negligible probability, the theorem's advertised regime would be falsified.
Extended reading notes
Core claim
The paper's central claim is Theorem 4.1: under lower bounds on the ambient dimension $D$, the curvature bound $\kappa$, and conditions $\sigma D\ge1$, $\sigma D^{20}\ge1/\kappa$, $\sigma\sqrt{D}\le c_1\tau$, the two-stage procedure in Algorithm 1 outputs a landmark $q_2$ with $d(q_2,\mathcal{M})\le C\sigma\sqrt{d(1+\kappa\,\mathrm{diam}(\mathcal{M})/\log D)}$ with probability at least $1-9e^{-c_2 d}$. The proof splits the average into a clean-signal average and a conditioned-noise average, controls the signal term by a curvature lemma, and controls the noise term through new subgaussian bounds on the probability $h(s)$ that a noisy point is accepted into the ball $B(q,R)$. The function $h(s)$ has a sharp phase transition at $s_\star^2=R^2-\sigma^2(D-3)$, and the paper shows the convolution of $h$ with the Gaussian noise density is subgaussian around that transition, which is what makes the high-noise regime tractable.
Load-bearing premise
The second refinement stage is proved only when $1/\kappa\ge C_5\sigma\sqrt{D\log D}$, a condition stricter than the headline $\sigma\sqrt{D}\le c_1\tau$ because the reach satisfies $\tau\le1/\kappa$; if the noise lies between those two levels, the claimed $\sigma\sqrt{d}$ second-round accuracy is not established.
Editorial extensions
If this is right
- Local averaging can serve as a preprocessing step that extends low-noise manifold-fitting guarantees to noise levels up to $\sigma\sqrt{D}\lesssim\tau$.
- Pairwise distances between clean signals can be estimated from local averages with error $O(\sigma d^{1/4}D^{1/4}(\log D)^{1/4})$, improving on the raw noisy-point distance in high dimensions.
- Replacing noisy samples by their local averages turns an $O(\sigma\sqrt{D})$ net for the manifold into an $O(\sigma(dD\log D)^{1/4})$ net, giving a discretization for downstream tasks such as denoising new samples.
- Because the algorithm only needs minibatches of accepted samples, it admits an online implementation and scales better with the ambient dimension than prior provable large-noise manifold-fitting algorithms.
Reading between the lines
- The advertised regime is narrower than the abstract suggests: the Stage II condition $1/\kappa\ge C_5\sigma\sqrt{D\log D}$ is stronger than $\sigma\sqrt{D}\le c_1\tau$ because the reach obeys $\tau\le1/\kappa$, so at fixed geometry the second-round bound requires noise about a factor $\sqrt{\log D}$ smaller.
- The phase-transition analysis of $h(s)$ is not tied to the two-round scheme; the same subgaussian bounds should transfer to mean shift, nonlocal-means patch averaging, and graph-Laplacian weight construction under Gaussian noise.
- The injected perturbation is present only to keep the landmark away from the manifold; replacing it with explicit regularization could yield a proof for arbitrarily many averaging rounds.
- The paper's own secant-bundle heuristic suggests the true limit may be $E[\Sigma]\approx\tau$ rather than $\sigma\sqrt{D}\approx\tau$; a direct test is whether trimmed or robust local averages can be analyzed at that higher noise boundary.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper analyzes a two-stage minibatch local averaging scheme for estimating a d-dimensional submanifold M ⊂ R^D from samples x_i = x_i,♮ + z_i with z_i ~ N(0, σ²I_D). In Stage I, averaging a minibatch of points in a ball of radius R1 around a noisy initialization is claimed to give d(q1, M) ≤ Cσ√(d(κdiam(M) + log D)) under σ√D ≤ c1τ, where τ is the reach. In Stage II, a Gaussian perturbation ϑ ~ N(0, σ²D^{-1/4}I) is injected, a second minibatch in a ball of radius R2 is averaged, and the refined bound d(q2, M) ≤ Cσ√(d(1 + κdiam(M)/log D)) is claimed under the additional hypothesis 1/κ ≥ C5σ√(D log D). The technical core is a subgaussian analysis of the grouping probability h(s), the probability that a noisy point falls in an extrinsic ball, obtained via incomplete-gamma-function bounds near its phase transition; this yields sharp bounds on the conditional noise expectation and the conditional intrinsic squared distance of grouped points, combined with Bernstein and vector-Hoeffding inequalities. The paper claims this is the first analysis of local averaging accuracy in the regime σ√D ≈ τ.
Significance. If fully established, the refined bound would be a valuable contribution to manifold estimation: the algorithm is simple, the radii and minibatch sizes are explicit functions of σ, d, D, κ and diam(M) rather than fitted quantities, and the phase-transition estimates for h(s) (Lemmas B.2–B.7) appear to be new, carefully proved, and likely reusable for tangent-space and curvature estimation. The paper is honest about its limitations (Section 6), including the lack of geometric-quantity estimation and the breakdown of the subgaussian tool when the landmark is very close to the manifold. The Stage I bound under σ√D ≤ c1τ is a solid contribution in its own right even if the Stage II regime claim must be narrowed. The main liabilities are the mismatch between the advertised and the proved Stage II regime, an unverified scaling condition in the Stage II proof of (A.9), and the reliance on two same-group preprints for load-bearing T_max bounds. These issues are fixable within the manuscript's scope, but they directly affect the paper's central novelty claim, so the present version is not ready.
major comments (3)
- [Abstract; §4, Theorem 4.1, Stage II] The abstract's headline claim is not supported by the hypotheses of Theorem 4.1. Stage II requires the additional hypothesis 1/κ ≥ C5σ√(D log D), and the reach satisfies τ ≤ 1/κ (a relation the paper itself uses, e.g., in Lemma C.2). When the reach is curvature-limited (τ ≈ 1/κ, as for a round sphere factor), this condition confines the refined bound to σ√D ≲ τ/√(log D), a factor √(log D) below the advertised 'noise comparable to the reach' regime; the abstract states the refined d(q̂, M) bound immediately after claiming the first analysis in the regime σ√D ≈ τ, conflating the Stage I regime (where σ√D ≤ c1τ is genuinely established) with the Stage II regime. I also could not locate any step in the Stage (c) proof that invokes 1/κ ≥ C5σ√(D log D): the verifications of (A.8)–(A.12) and the bounds (A.57)–(A.108) use only σ√D ≤ c1τ, σD^{20} ≥ 1/κ, and D^{1/12} ≥ C1 max{κ̄²d, κdiam(M)d}. If the condition is unnecessary, it should be deleted or weakened to the requirement 1/κ ≥ CσD^{5/24} forced by (A.10), which would restore the advertised regime and make the abstract accurate; if it is necessary, the exact step using it must be identified. I note that the stress-test concern that condition (A.12) fails at the advertised boundary does not land: the RHS of (A.12) is c4š²⋆, of order σ²D^{3/4} in Stage II, so the term σ²√(D log D) is negligible (ratio √(log D)/D^{1/4} → 0) rather than divergent; the regime gap comes from the stated hypothesis and from (A.9), not from (A.12).
- [Appendix A, Stage (c), verification of (A.9)] The verification of the second-round conditions of Theorems B.1 and C.1 is incomplete, and the step verifying (A.9) has a genuine scaling gap. In Stage II, s⋆² = σ²(D^{3/4} + 2C D^{5/12}), so (A.9) requires C2(log D + κddiam(M) + d log(1/(κs⋆,∥)))σ²D^{2/3} ≤ s⋆², i.e., the bracket must be O(D^{1/12}). The text bounds the third bracket term only by C d log D (via σD^{20} ≥ 1/κ and s⋆,∥ ≥ CσD^{5/24}), while the hypothesis D^{1/12} ≥ C1κ̄²d, κ̄ ≥ 1, gives only d ≤ D^{1/12}/C1; these do not imply d log D ≤ cD^{1/12}. Along the maximal allowed choice d ≈ D^{1/12}/(C1κ̄²), the bracket is Θ(d log D) = Θ(D^{1/12} log D), exceeding the required size by the factor Θ(log D). This is not merely a constants issue, since the discrepancy is in the scaling with D. The subsequent sentence 'Similar to round 1' cannot discharge (A.10)–(A.12) either, because their right-hand sides in the second round scale as σ²D^{3/4} rather than σ²D as in the first round. The fix is to add an explicit large-D assumption such as D^{1/12} ≥ C8(d log D + log D + κdiam(M)d) and to spell out the verification of (A.10)–(A.12) for the second round.
- [Appendix A, Eqs. (A.64)–(A.69), and Lemma D.2] The refined bound is not self-contained at a load-bearing point. The lower bound on d(q1, M) in (A.70)–(A.84), which sets the scale of the second-round radius, uses the bound Tmax,ϑ ≤ Cκ̄σD^{-1/2}√d from Lemma A.4 of [10] and the statement 'which along with Lemma B.5 from [10]' at (A.67)–(A.68); Lemma D.2 similarly uses Theorem 3 of [31] for E[Tmax] ≤ Cκ̄σ√d. References [10] (arXiv:2502.16819) and [31] (TPoPT) are same-group preprints whose proofs are not reproduced in this manuscript. Since the Stage II claim is the paper's central contribution, the authors should either prove these short statements (which appear to follow from the ε-net/covering arguments already present in Section B) or cite refereed versions; otherwise the 'first analysis' claim rests on unpublished work.
minor comments (7)
- [Throughout] There are several typos: 'decribe' in §5.1, 'denoiing' in the acknowledgments, 'signficantly' in §4 after Eq. (4.2), and 'optimziation' in the title of [31].
- [Algorithm 1 / §3] The injected perturbation ϑ ∼ N(0, σ²D^{-1/4}I) is introduced without comment in Algorithm 1; the reader only learns in footnote 6 that it is a purely technical device. Since the paper motivates the method by practical simplicity, the main text should flag this at the algorithm.
- [Figure 1 caption] The caption says that d(q2, M) is bounded by σ√d 'regardless of ambient dimension D'; the theorem's bound is Cσ√(d(1 + κdiam(M)/log D)), which is independent of D only in scaling, so the caption should be reworded.
- [Lemma C.2 / problem setup] The relation τ ≤ 1/κ between the reach and the curvature bound is used implicitly (Lemma C.2) but never stated; it should be stated and cited near the problem setup, especially since the Stage II regime discussion depends on it.
- [Appendix D, Lemma D.5] The proof of Stirling's formula via the de Moivre–Laplace theorem spans several pages; a standard reference for Stirling's bound would shorten the appendix considerably.
- [References] Reference [31] is cited as 'preprint, 2023' without an arXiv identifier; please provide one for verifiability.
- [§5.1, Eqs. (5.2)–(5.3)] These sketched bounds appear to refer to the second round, but the round is not specified; the round-1 noise-average bound (A.29) lacks the D^{-1/8} factor that appears in (5.3), so an explicit statement would prevent confusion.
Circularity Check
No significant circularity: the two-stage bound is derived from explicit assumptions, and the self-cited T_max lemmas are auxiliary, parameter-free concentration bounds.
full rationale
The paper is a self-contained probabilistic analysis. Algorithm 1's radii and minibatch sizes are explicit functions of σ, D, d, κ, and diam(M), and Theorem 4.1's bound is obtained by combining subgaussian estimates for the chi-square grouping probability h(s), the signal-error lemma, and Hoeffding/Bernstein concentration inequalities. None of these steps fits a parameter to the target quantity, and no 'prediction' is an input by construction. The self-citations to [10] and [31] supply sup-over-tangent-bundle concentration bounds for T_max used in the Stage II lower bound; these are auxiliary, parameter-free lemmas about Gaussian maxima over tangent spaces, not assertions of the local-averaging accuracy itself, and they do not include the target result among their assumptions. The additional Stage II hypothesis 1/κ ≥ C5σ√(D log D) is a disclosed proof condition; whether it conflicts with the abstract's advertised 'σ√D ≈ τ' regime is a correctness/scope concern, not a circularity. No load-bearing step reduces to its own input by definition or by fitting.
Assumptions & free parameters
free parameters (2)
- Neighborhood radii R1, R2 =
R1²=σ²(2D-d-3)+C3κ̄²σ²d+3C3κ̄σ²√(Dd); R2²=σ²(D-3+D^{3/4}+2C7D^{5/12})
- Injected noise variance for second round =
σ² D^{-1/4}
assumptions (5)
- domain assumption M is a compact, connected, geodesically complete d-dimensional submanifold of R^D with extrinsic curvature bounded by κ and reach τ.
- domain assumption Clean signals x^♮ are uniformly sampled from M; noise z_i are iid N(0,σ²I).
- domain assumption Parameter conditions: σD≥1, σD^20≥1/κ, σ√D≤c1τ, D^{1/12}≥C max{κ̄²d, κdiam(M)d}, and for Stage II 1/κ≥C5σ√(D log D).
- domain assumption κdiam(M) ≥ 1.
- ad hoc to paper Lemma A.4 from [10] and Theorem 3 from [31] hold, bounding T_max over tangent spaces of the manifold.
Cite this review
Pith. "Pith review of Local Averaging Accurately Distills Manifold Structure From Noisy Data." pith.science (2026). https://pith.science/paper/EPO734PK
@misc{pith2026250618761,
author = {Pith},
title = {Pith review of: Local Averaging Accurately Distills Manifold Structure From Noisy Data},
year = {2026},
howpublished = {\url{https://pith.science/paper/EPO734PK}},
note = {Machine review of arXiv:2506.18761}
}
abstract
High-dimensional data are ubiquitous, with examples ranging from natural images to scientific datasets, and often reside near low-dimensional manifolds. Leveraging this geometric structure is vital for downstream tasks, including signal denoising, reconstruction, and generation. However, in practice, the manifold is typically unknown and only noisy samples are available. A fundamental approach to uncovering the manifold structure is local averaging, which is a cornerstone of state-of-the-art provable methods for manifold fitting and denoising. However, to the best of our knowledge, there are no works that rigorously analyze the accuracy of local averaging in a manifold setting in high-noise regimes. In this work, we provide theoretical analyses of a two-round mini-batch local averaging method applied to noisy samples drawn from a $d$-dimensional manifold $\mathcal M \subset \mathbb{R}^D$, under a relatively high-noise regime where the noise size is comparable to the reach $\tau$. We show that with high probability, the averaged point $\hat{\mathbf q}$ achieves the bound $d(\hat{\mathbf q}, \mathcal M) \leq \sigma \sqrt{d\left(1+\frac{\kappa\mathrm{diam}(\mathcal {M})}{\log(D)}\right)}$, where $\sigma, \mathrm{diam(\mathcal M)},\kappa$ denote the standard deviation of the Gaussian noise, manifold's diameter and a bound on its extrinsic curvature, respectively. This is the first analysis of local averaging accuracy over the manifold in the relatively high noise regime where $\sigma \sqrt{D} \approx \tau$. The proposed method can serve as a preprocessing step for a wide range of provable methods designed for lower-noise regimes. Additionally, our framework can provide a theoretical foundation for a broad spectrum of denoising and dimensionality reduction methods that rely on local averaging techniques.
Figures
Reference graph
Works this paper leans on
-
[7]
Yariv Aizenbud and Barak Sober. Estimation of local geometric structure on manifolds from noisy data.Journal of Machine Learning Research, 26(64):1–89, 2025
work page 2025
-
[10]
Fast, Accurate Manifold Denoising by Tunneling Riemannian Optimization
Shiyu Wang et al. Fast, accurate manifold denoising by tunneling riemannian opti- mization, 2025. arXiv preprint arXiv:2502.16819
work page Pith review arXiv 2025
-
[31]
Tpopt: Efficienttrainabletemplateoptimziationonlow-dimensionalmanifolds
Jingkai Yan, Shiyu Wang, Xinyu Rain Wei, Zsuzsa Marka, Szabolcs Marka, and John Wright. Tpopt: Efficienttrainabletemplateoptimziationonlow-dimensionalmanifolds. preprint, 2023
work page 2023
-
[1]
Keinosuke Fukunaga and Larry Hostetler. The estimation of the gradient of a density function, with applications in pattern recognition.IEEE Transactions on information theory, 21(1):32–40, 1975
work page 1975
-
[2]
A non-local algorithm for image de- noising
Antoni Buades, Bartomeu Coll, and J-M Morel. A non-local algorithm for image de- noising. In2005IEEEcomputersocietyconferenceoncomputervisionandpatternrecognition (CVPR’05), volume 2, pages 60–65. Ieee, 2005
work page 2005
-
[3]
Antoni Buades, Bartomeu Coll, and Jean-Michel Morel. Non-local means denoising. Image Processing On Line, 1:208–212, 2011
work page 2011
-
[4]
Roweis and Lawrence K
Sam T. Roweis and Lawrence K. Saul. Nonlinear dimensionality reduction by locally linear embedding.Science, 290(5500):2323–2326, 2000
2000
-
[5]
Zhigang Yao, Jiaji Su, Bingjie Li, and Shing-Tung Yau. Manifold fitting, 2023
work page 2023
Show all 36 references
-
[6]
Manifold fitting under unbounded noise.Journal of Machine Learning Research, 26(45):1–55, 2025
Zhigang Yao and Yuqing Xia. Manifold fitting under unbounded noise.Journal of Machine Learning Research, 26(45):1–55, 2025
2025
-
[8]
Towardsatheoreticalfoundationforlaplacian-based manifold methods.Journal of Computer and System Sciences, 74(8):1289–1308, 2008
MikhailBelkinandParthaNiyogi. Towardsatheoreticalfoundationforlaplacian-based manifold methods.Journal of Computer and System Sciences, 74(8):1289–1308, 2008
2008
-
[9]
Local regularization of noisy point clouds: Improved global geometric estimates and data analysis.Journal of Machine Learning Research, 20(136):1–37, 2019
Nicolás García Trillos, Daniel Sanz-Alonso, and Ruiyi Yang. Local regularization of noisy point clouds: Improved global geometric estimates and data analysis.Journal of Machine Learning Research, 20(136):1–37, 2019
2019
-
[11]
Eine methode zur auffindung von gruppen.Biometrische Zeitschrift, 6(1):47– 48, 1964
P Schnell. Eine methode zur auffindung von gruppen.Biometrische Zeitschrift, 6(1):47– 48, 1964
1964
-
[12]
Nonparametric ridge estimation
Christopher R Genovese, Marco Perone-Pacifico, Isabella Verdinelli, and Larry Wasser- man. Nonparametric ridge estimation. 2014
2014
-
[13]
Mean shift analysis and applications
Dorin Comaniciu and Peter Meer. Mean shift analysis and applications. InProceedings of the seventh IEEE international conference on computer vision, volume 2, pages 1197–1203. IEEE, 1999
1999
-
[14]
On convergence of epanechnikov mean shift
Kejun Huang, Xiao Fu, and Nicholas Sidiropoulos. On convergence of epanechnikov mean shift. InProceedings of the AAAI Conference on Artificial Intelligence, volume 32, 2018
2018
-
[15]
Gaussianmean-shiftisanemalgorithm.IEEETransactions on Pattern Analysis and Machine Intelligence, 29(5):767–776, 2007
MiguelACarreira-Perpinan. Gaussianmean-shiftisanemalgorithm.IEEETransactions on Pattern Analysis and Machine Intelligence, 29(5):767–776, 2007. 10
2007
-
[16]
Convergence analysis of mean shift.IEEE transactions on pattern analysis and machine intelligence, 2024
Ryoya Yamasaki and Toshiyuki Tanaka. Convergence analysis of mean shift.IEEE transactions on pattern analysis and machine intelligence, 2024
2024
-
[17]
Nonasymptotic rates for manifold, tangent space and curvature estimation
Eddie Aamari and Clément Levrard. Nonasymptotic rates for manifold, tangent space and curvature estimation. 2019
2019
-
[18]
Manifold blurring mean shift algo- rithms for manifold denoising
Weiran Wang and Miguel A Carreira-Perpinán. Manifold blurring mean shift algo- rithms for manifold denoising. In2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pages 1759–1766. IEEE, 2010
2010
-
[19]
Manifold learning using kernel density estimation and local principal components analysis.arXiv preprint arXiv:1709.03615, 2017
Kitty Mohammed and Hariharan Narayanan. Manifold learning using kernel density estimation and local principal components analysis.arXiv preprint arXiv:1709.03615, 2017
2017 arXiv
-
[20]
Smooth manifold reconstruction from noisy and non-uniform approximation with guarantees.Computational Geometry, 40(2):156–170, 2008
Frédéric Chazal and André Lieutier. Smooth manifold reconstruction from noisy and non-uniform approximation with guarantees.Computational Geometry, 40(2):156–170, 2008
2008
-
[21]
Minimax manifold estimation.Journal of machine learning research, 13:1263–1291, 2012
Christopher R Genovese, Marco Perone Pacifico, Verdinelli Isabella, Larry Wasserman, et al. Minimax manifold estimation.Journal of machine learning research, 13:1263–1291, 2012
2012
-
[22]
Finding the homology of submanifolds with high confidence from random samples.Discrete & Computational Geometry, 39:419–441, 2008
Partha Niyogi, Stephen Smale, and Shmuel Weinberger. Finding the homology of submanifolds with high confidence from random samples.Discrete & Computational Geometry, 39:419–441, 2008
2008
-
[23]
Fitting a manifold of large reach to noisy data.arXiv preprint arXiv:1910.05084, 2019
Charles Fefferman, Sergei Ivanov, Matti Lassas, and Hariharan Narayanan. Fitting a manifold of large reach to noisy data.arXiv preprint arXiv:1910.05084, 2019
1910 arXiv
-
[24]
Tenenbaum, Vin de Silva, and John C
Joshua B. Tenenbaum, Vin de Silva, and John C. Langford. A global geometric frame- work for nonlinear dimensionality reduction.Science, 290(5500):2319–2323, 2000
2000
-
[25]
Manifold denoising.Advances in neural information processing systems, 19, 2006
Matthias Hein and Markus Maier. Manifold denoising.Advances in neural information processing systems, 19, 2006
2006
-
[26]
Fitting a manifold to data in the presence of large noise.arXiv preprint arXiv:2312.10598, 2023
Charles Fefferman, Sergei Ivanov, Matti Lassas, and Hariharan Narayanan. Fitting a manifold to data in the presence of large noise.arXiv preprint arXiv:2312.10598, 2023
2023 arXiv
-
[27]
Cambridge university press, 2018
Roman Vershynin.High-dimensional probability: An introduction with applications in data science, volume 47. Cambridge university press, 2018
2018
-
[28]
Wainwright.High-Dimensional Statistics: A Non-Asymptotic Viewpoint
Martin J. Wainwright.High-Dimensional Statistics: A Non-Asymptotic Viewpoint. Cam- bridge University Press, Cambridge, UK; New York, NY, 2019
2019
-
[29]
Existence, uniqueness and regularity of the projection onto differentiable manifolds.Annals of global analysis and geometry, 60(3):559–587, 2021
Gunther Leobacher and Alexander Steinicke. Existence, uniqueness and regularity of the projection onto differentiable manifolds.Annals of global analysis and geometry, 60(3):559–587, 2021
2021
-
[30]
Adaptive estimation of a quadratic functional by model selection.Annals of statistics, pages 1302–1338, 2000
Beatrice Laurent and Pascal Massart. Adaptive estimation of a quadratic functional by model selection.Annals of statistics, pages 1302–1338, 2000
2000
-
[32]
1 Nmb NmbX ℓ=1 zℓ−E
Joel A. Tropp. User-friendly tail bounds for sums of random matrices.Foundations of Computational Mathematics, 12(4):389–434, 2012. 11 A Analyzing Landmarking We are motivated by the following minibatch gradient algorithm for landmarking: Algorithm 2MinibatchSGDLandmarking Inp...
2012
-
[33]
d(x,M)1 ∥z∥≤ 1 6τ # ≥E
Any critical point¯x♮ of this function satisfies q− ¯x♮∈N ¯x♮M(C.34) I.e., if we consider the normal mapping(x♮,η)7→N(x ♮,η) =x ♮ +η, then we have N( ¯x♮,q− ¯x♮) =q=N(q ♮,q−q ♮).(C.35) 40 Since the normal map is injective on M×int BD(0,τ) (C.36) this implies that∥q− ¯x♮∥≥τ. Ca...
-
[34]
Observe that Cn−Cn+1 =−log(n+ 1)−(n+ 1
log(n)+ n. Observe that Cn−Cn+1 =−log(n+ 1)−(n+ 1
-
[35]
log(n+ 1)−1 =−1 + (n+ 1
-
[36]
1 N NX ℓ=1 zℓ−E
log(1 + 1 n) =−1 + ( 1 2n+ 1 ) log(2n+ 2 2n ) 45 We lett= 1 2n+1, then, we have Cn−Cn+1 =−1 + ( 1 2t) log(1 +t 1−t ) =−1 + ( 1 2t)(−2)(t+ t3 3 + t5 5 +...) = X k≥1 t2k 2k+ 1 Since we have both the following lower and upper bound, X k≥1 t2k 2k+ 1 ≥ t2 3 = 1 12n2 + 12n+ 3 ≥ 1 12...
Reviewed August 15, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.