pith. sign in

arxiv: 2606.06483 · v1 · pith:X3MR44NZnew · submitted 2026-06-04 · 🧮 math.ST · math.OC· math.SP· stat.ME· stat.TH

Statistically and Computationally Optimal Estimation and Inference of Common Subspaces

Joshua Agterberg This is my paper

Pith reviewed 2026-06-27 23:09 UTC · model grok-4.3

classification 🧮 math.ST math.OCmath.SPstat.MEstat.TH
keywords common subspacesignal-to-noise ratio regimesstatistical and computational limitsprojected gradient descentasymptotic normalityadaptive confidence intervalslow-rank matricessin Theta distance
0
0 comments X

The pith

Adaptive inference on common subspace error is impossible below a higher SNR threshold than estimation.

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

The paper examines the common subspace model in which multiple symmetric low-rank matrices share one subspace and are observed with additive noise. It partitions the signal-to-noise ratio into four asymptotic regimes and shows that a projected gradient descent estimator, started from a spectral sum-of-squares initializer, attains the minimax sin Theta rate precisely when the SNR enters the strong estimation regime. The same estimator is shown to be asymptotically normal only once the SNR reaches the still-higher strong inference regime, which permits construction of adaptively minimax-optimal confidence intervals; below that regime such adaptive intervals are proved impossible even though estimation itself remains computationally feasible.

Core claim

In the common subspace model the signal-to-noise ratio is partitioned into weak, moderate, strong estimation, and strong inference regimes; projected gradient descent initialized by spectral sum of squares attains the optimal sin Theta error rate in the strong estimation regime, the estimator is asymptotically Gaussian in the strong inference regime and yields adaptively optimal confidence intervals there, while adaptive confidence intervals are information-theoretically impossible in all lower regimes.

What carries the argument

Four SNR regimes (weak, moderate, strong estimation, strong inference) that mark the boundaries between statistical and computational feasibility for estimation and for inference of the shared subspace.

If this is right

  • Projected gradient descent initialized by spectral sum of squares attains the optimal sin Theta error rate once the SNR enters the strong estimation regime.
  • The estimator is asymptotically Gaussian precisely in the strong inference regime.
  • Confidence intervals derived from the limiting distribution are adaptively minimax optimal in the strong inference regime.
  • Adaptive confidence intervals are information-theoretically impossible in every regime below the strong inference threshold.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The separation between the estimation and inference thresholds indicates that inference tasks can impose stricter information-theoretic requirements than estimation even in the same model.
  • Analogous gaps between estimation and inference limits may exist in related multi-view or tensor models with shared factors.
  • Applied work that uses estimated subspaces for downstream inference must check whether the observed SNR exceeds the higher inference threshold.
  • The results motivate distinguishing computational feasibility of estimation from statistical feasibility of inference when designing procedures for high-dimensional shared-structure problems.

Load-bearing premise

The data consist of symmetric low-rank matrices sharing exactly the same common subspace and each perturbed by independent noise, with matrix dimensions and rank growing according to the four asymptotic SNR regimes.

What would settle it

An explicit construction of adaptive confidence intervals for the sin Theta distance that attain the minimax rate at an SNR strictly below the strong inference threshold would refute the impossibility claim.

Figures

Figures reproduced from arXiv: 2606.06483 by Joshua Agterberg.

Figure 1
Figure 1. Figure 1: Simulation of proposed projected gradient descent algorithm (solid lines) versus the spectral [PITH_FULL_IMAGE:figures/full_fig_p014_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Resulting embedding of trade data Dimension 1 Dimension 2 Dimension 3 Dimension 4 Dimension 5 [PITH_FULL_IMAGE:figures/full_fig_p015_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Percentile of each column of Ubt for each country, with warmer colors associated to larger values and cooler colors associated to smaller values. 7 Discussion We have considered the statistical and computational limits for estimation of U with respect to the Frobenius sin Θ distance, and the statistical limits for adaptive inference for the error ∥ sin Θ(Ub , U)∥ 2 F . In practice, these results imply that… view at source ↗
read the original abstract

Given multiple data matrices, many problems in statistics and data science rely on estimating a common subspace that captures certain structure shared by all the data matrices. In this paper we investigate the statistical and computational limits for the common subspace model in which one observes a collection of symmetric low-rank matrices perturbed by noise, where each low-rank matrix shares the same common subspace. Our main results identify several regimes of the signal-to-noise ratio (SNR) such that estimation and inference are statistically or computationally optimal, and we refer to these regimes as weak SNR, moderate SNR, strong estimation SNR, and strong inference SNR. First, we propose an estimator based on projected gradient descent initialized via spectral sum of squares and show that it achieves the optimal $\sin\Theta$ error rate under strong estimation SNR. These results are complemented by both statistical and computational lower bounds identifying the weak and moderate estimation SNR regimes. Next, we turn to statistical inference for the $\sin\Theta$ distance itself, and we show that our estimator has an asymptotically Gaussian distribution in the strong inference SNR regime. Based on this limiting result we propose confidence intervals and show that they are adaptively minimax optimal in the strong inference SNR regime, where adaptivity is measured in terms of the SNR. Finally, we show that adaptive confidence intervals are information-theoretically impossible below the strong inference SNR regime. Consequently, our results unveil a novel phenomenon: despite the SNR being ``above'' the computational limit for estimation, adaptive statistical inference may still be information-theoretically impossible.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 2 minor

Summary. The paper studies the common subspace model in which multiple symmetric low-rank matrices, each perturbed by independent noise, share an identical common subspace. It identifies four SNR regimes (weak, moderate, strong estimation, and strong inference) and derives matching upper and lower bounds on estimation error, an estimator based on spectral initialization followed by projected gradient descent that attains the optimal sin Θ rate under strong estimation SNR, asymptotic normality of the estimator under strong inference SNR, adaptively minimax-optimal confidence intervals in that regime, and an information-theoretic impossibility result for adaptive CIs below the strong inference threshold.

Significance. If the derivations hold, the work supplies a fine-grained separation of statistical, computational, and adaptive-inference thresholds for a standard multi-matrix subspace model. The explicit construction of an estimator achieving the information-theoretic rate together with matching lower bounds, plus the demonstration that adaptive inference can remain impossible even when estimation is computationally feasible, constitutes a substantive contribution to the literature on high-dimensional subspace estimation and inference.

minor comments (2)
  1. [Abstract] Abstract: the four SNR regimes are named but their precise parameter thresholds (in terms of dimension, rank, and noise level) are not stated; adding one sentence that locates each regime would improve immediate readability.
  2. The model statement (symmetric matrices, exact common subspace, independent noise) is standard, but the precise growth rates assumed for p, r, and n in the four asymptotic regimes should be collected in one displayed assumption block for reference.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive evaluation and recommendation of minor revision. No major comments appear in the report, so we have no points requiring point-by-point rebuttal. We will incorporate any minor suggestions in the revised version.

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The derivation chain separates into independent components: an estimator (projected gradient descent from spectral initialization) whose sin Θ error rate is shown to match statistical and computational lower bounds derived from the model assumptions (symmetric low-rank matrices with independent noise sharing a common subspace); asymptotic normality of the estimator in the strong inference SNR regime; construction of adaptive CIs that achieve minimax optimality there; and a separate information-theoretic argument establishing impossibility of adaptive CIs below that threshold. None of these steps reduce by construction to fitted inputs, self-definitions, or load-bearing self-citations; all bounds and limits are obtained from the explicitly stated asymptotic regimes and noise model without circular renaming or imported uniqueness theorems.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review; no explicit free parameters, axioms, or invented entities can be extracted. The model implicitly relies on standard additive noise and exact common-subspace assumptions typical in the field.

pith-pipeline@v0.9.1-grok · 5806 in / 1158 out tokens · 21811 ms · 2026-06-27T23:09:29.534929+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

116 extracted references · 103 canonical work pages

  1. [1]

    Journal of Machine Learning Research , author =

    Inference for. Journal of Machine Learning Research , author =. 2021 , pages =

    2021

  2. [2]

    , month = feb, year =

    Luo, Yuetian and Zhang, Anru R. , month = feb, year =. Tensor clustering with planted structures:. The Annals of Statistics , publisher =. doi:10.1214/21-AOS2123 , abstract =

    work page doi:10.1214/21-aos2123

  3. [3]

    and Zhu, Zihan , month = oct, year =

    Lei, Jing and Zhang, Anru R. and Zhu, Zihan , month = oct, year =. Computational and statistical thresholds in multi-layer stochastic block models , volume =. The Annals of Statistics , publisher =. doi:10.1214/24-AOS2441 , abstract =

    work page doi:10.1214/24-aos2441

  4. [4]

    The Annals of Statistics , author =

    Xia, Dong and Zhang, Anru R. and Zhou, Yuchen , month = apr, year =. Inference for low-rank tensors—no need to debias , volume =. The Annals of Statistics , publisher =. doi:10.1214/21-AOS2146 , abstract =

    work page doi:10.1214/21-aos2146

  5. [5]

    IEEE Transactions on Information Theory , author =

    Tensor. IEEE Transactions on Information Theory , author =. 2018 , keywords =. doi:10.1109/TIT.2018.2841377 , abstract =

    work page doi:10.1109/tit.2018.2841377 2018

  6. [6]

    Optimal estimation and computational limit of low-rank

    Lyu, Zhongyuan and Xia, Dong , month = apr, year =. Optimal estimation and computational limit of low-rank. The Annals of Statistics , publisher =. doi:10.1214/23-AOS2264 , abstract =

    work page doi:10.1214/23-aos2264

  7. [7]

    IEEE Transactions on Information Theory , author =

    Bias-. IEEE Transactions on Information Theory , author =. 2024 , keywords =. doi:10.1109/TIT.2024.3471953 , abstract =

    work page doi:10.1109/tit.2024.3471953 2024

  8. [8]

    Journal of the American Statistical Association , volume =

    Lei, Jing and and Lin, Kevin Z. , month = oct, year =. Bias-. Journal of the American Statistical Association , publisher =. doi:10.1080/01621459.2022.2054817 , abstract =

    work page doi:10.1080/01621459.2022.2054817 2022

  9. [9]

    Statistical

    Agterberg, Joshua and Zhang, Anru , note =. Statistical. The Annals of Statistics, To Appear , publisher =. doi:10.48550/arXiv.2410.06381 , abstract =

    work page doi:10.48550/arxiv.2410.06381

  10. [10]

    IEEE Transactions on Signal Processing , author =

    Nonconvex. IEEE Transactions on Signal Processing , author =. 2019 , keywords =. doi:10.1109/TSP.2019.2937282 , abstract =

    work page doi:10.1109/tsp.2019.2937282 2019

  11. [11]

    Foundations and Trends® in Machine Learning , author =

    Chen, Yuxin and Chi, Yuejie and Fan, Jianqing and Ma, Cong , month = oct, year =. Spectral. Foundations and Trends® in Machine Learning , publisher =. doi:10.1561/2200000079 , abstract =

    work page doi:10.1561/2200000079

  12. [12]

    , month = feb, year =

    Han, Rungang and Willett, Rebecca and Zhang, Anru R. , month = feb, year =. An optimal statistical and computational framework for generalized tensor estimation , volume =. The Annals of Statistics , publisher =. doi:10.1214/21-AOS2061 , abstract =

    work page doi:10.1214/21-aos2061

  13. [13]

    , month = jun, year =

    Schramm, Tselil and Wein, Alexander S. , month = jun, year =. Computational barriers to estimation from low-degree polynomials , volume =. The Annals of Statistics , publisher =. doi:10.1214/22-AOS2179 , abstract =

    work page doi:10.1214/22-aos2179

  14. [14]

    Tony and Guo, Zijian and Ma, Rong , month = apr, year =

    Chen, Elynn Y. and and Fan, Jianqing , month = apr, year =. Statistical. Journal of the American Statistical Association , publisher =. doi:10.1080/01621459.2021.1970569 , number =

    work page doi:10.1080/01621459.2021.1970569 2021

  15. [15]

    Vincent and Chen, Yuxin , month = apr, year =

    Cai, Changxiao and Li, Gen and Chi, Yuejie and Poor, H. Vincent and Chen, Yuxin , month = apr, year =. Subspace estimation from unbalanced and incomplete data matrices: \. The Annals of Statistics , publisher =. doi:10.1214/20-AOS1986 , abstract =

    work page doi:10.1214/20-aos1986

  16. [16]

    and Cai, T

    Zhang, Anru R. and Cai, T. Tony and Wu, Yihong , month = feb, year =. Heteroskedastic. The Annals of Statistics , publisher =. doi:10.1214/21-AOS2074 , abstract =

    work page doi:10.1214/21-aos2074

  17. [17]

    Community

    Jing, Bing-Yi and Li, Ting and Lyu, Zhongyuan and Xia, Dong , year =. Community. The Annals of Statistics , publisher =

  18. [18]

    The Annals of Statistics , author =

    Zhou, Yuchen and Chen, Yuxin , month = feb, year =. Deflated. The Annals of Statistics , publisher =. doi:10.1214/24-AOS2456 , abstract =

    work page doi:10.1214/24-aos2456

  19. [19]

    , month = dec, year =

    Luo, Yuetian and Zhang, Anru R. , month = dec, year =. Tensor-on-tensor regression:. The Annals of Statistics , publisher =. doi:10.1214/24-AOS2396 , abstract =

    work page doi:10.1214/24-aos2396

  20. [20]

    Zhang, Anru and Han, Rungang , month = oct, year =. Optimal. Journal of the American Statistical Association , publisher =

  21. [21]

    Journal of the American Statistical Association , volume =

    Cai, Jian-Feng and , Jingyang, Li and and Xia, Dong , month = oct, year =. Generalized. Journal of the American Statistical Association , publisher =. doi:10.1080/01621459.2022.2063131 , abstract =

    work page doi:10.1080/01621459.2022.2063131 2022

  22. [22]

    Journal of the Royal Statistical Society Series B: Statistical Methodology , author =

    Exact. Journal of the Royal Statistical Society Series B: Statistical Methodology , author =. 2022 , pages =. doi:10.1111/rssb.12547 , abstract =

    work page doi:10.1111/rssb.12547 2022

  23. [23]

    Journal of Machine Learning Research , author =

    Higher-. Journal of Machine Learning Research , author =. 2023 , pages =

    2023

  24. [24]

    IEEE Transactions on Information Theory , author =

    Fundamental. IEEE Transactions on Information Theory , author =. 2024 , keywords =. doi:10.1109/TIT.2024.3425581 , abstract =

    work page doi:10.1109/tit.2024.3425581 2024

  25. [25]

    Lyu, Zhongyuan and Li, Ting and Xia, Dong , month = nov, year =. Optimal. doi:10.48550/arXiv.2311.15598 , abstract =

    work page doi:10.48550/arxiv.2311.15598

  26. [26]

    Tensors in

    Auddy, Arnab and Xia, Dong and Yuan, Ming , month = mar, year =. Tensors in. Annual Review of Statistics and Its Application , publisher =. doi:10.1146/annurev-statistics-112723-034548 , abstract =

    work page doi:10.1146/annurev-statistics-112723-034548

  27. [27]

    Heteroskedastic

    Zhou, Yuchen and Chen, Yuxin , month = nov, year =. Heteroskedastic. doi:10.48550/arXiv.2311.02306 , abstract =

    work page doi:10.48550/arxiv.2311.02306

  28. [28]

    The Annals of Statistics , author =

    Yan, Yuling and Chen, Yuxin and Fan, Jianqing , month = apr, year =. Inference for heteroskedastic. The Annals of Statistics , publisher =. doi:10.1214/24-AOS2366 , abstract =

    work page doi:10.1214/24-aos2366

  29. [29]

    Global and individualized community detection in inhomogeneous multilayer networks , volume =

    Chen, Shuxiao and Liu, Sifan and Ma, Zongming , month = oct, year =. Global and individualized community detection in inhomogeneous multilayer networks , volume =. The Annals of Statistics , publisher =. doi:10.1214/22-AOS2202 , abstract =

    work page doi:10.1214/22-aos2202

  30. [30]

    Bernoulli , author =

    Du, Xinjie and Tang, Minh , month = nov, year =. Hypothesis testing for equality of latent positions in random graphs , volume =. Bernoulli , publisher =. doi:10.3150/22-BEJ1581 , abstract =

    work page doi:10.3150/22-bej1581

  31. [31]

    Journal of the Royal Statistical Society Series B: Statistical Methodology , author =

    Simple:. Journal of the Royal Statistical Society Series B: Statistical Methodology , author =. 2022 , pages =. doi:10.1111/rssb.12505 , abstract =

    work page doi:10.1111/rssb.12505 2022

  32. [32]

    Spectral

    Huang, Sihan and Weng, Haolei and Feng, Yang , month = jul, year =. Spectral. Journal of Computational and Graphical Statistics , publisher =

  33. [33]

    Ke, Zheng Tracy and and Wang, Jingming , note =. Optimal. Journal of the American Statistical Association , publisher =. doi:10.1080/01621459.2024.2388903 , abstract =

    work page doi:10.1080/01621459.2024.2388903 2024

  34. [34]

    Consistency of spectral clustering in stochastic block models , volume =

    Lei, Jing and Rinaldo, Alessandro , month = feb, year =. Consistency of spectral clustering in stochastic block models , volume =. The Annals of Statistics , publisher =. doi:10.1214/14-AOS1274 , abstract =

    work page doi:10.1214/14-aos1274

  35. [35]

    Biometrika , author =

    Consistent community detection in multi-layer network data , volume =. Biometrika , author =. 2020 , pages =. doi:10.1093/biomet/asz068 , abstract =

    work page doi:10.1093/biomet/asz068 2020

  36. [36]

    Journal of the American Statistical Association116(536), 1983–1993 (2021) https://doi.org/10

    Mao, Xueyu and , Purnamrita, Sarkar and and Chakrabarti, Deepayan , month = oct, year =. Estimating. Journal of the American Statistical Association , publisher =. doi:10.1080/01621459.2020.1751645 , abstract =

    work page doi:10.1080/01621459.2020.1751645 2020

  37. [37]

    Journal of Multivariate Analysis , author =

    Sparse subspace clustering in diverse multiplex network model , volume =. Journal of Multivariate Analysis , author =. 2024 , keywords =. doi:10.1016/j.jmva.2024.105333 , abstract =

    work page doi:10.1016/j.jmva.2024.105333 2024

  38. [38]

    Spectral and matrix factorization methods for consistent community detection in multi-layer networks , volume =

    Paul, Subhadeep and Chen, Yuguo , month = feb, year =. Spectral and matrix factorization methods for consistent community detection in multi-layer networks , volume =. The Annals of Statistics , publisher =. doi:10.1214/18-AOS1800 , abstract =

    work page doi:10.1214/18-aos1800

  39. [39]

    Sankhya A , author =

    Null. Sankhya A , author =. 2022 , keywords =. doi:10.1007/s13171-021-00257-0 , abstract =

    work page doi:10.1007/s13171-021-00257-0 2022

  40. [40]

    Spectral clustering and the high-dimensional stochastic blockmodel , volume =

    Rohe, Karl and Chatterjee, Sourav and Yu, Bin , month = aug, year =. Spectral clustering and the high-dimensional stochastic blockmodel , volume =. The Annals of Statistics , publisher =. doi:10.1214/11-AOS887 , abstract =

    work page doi:10.1214/11-aos887

  41. [41]

    Journal of the Royal Statistical Society Series B: Statistical Methodology , author =

    A. Journal of the Royal Statistical Society Series B: Statistical Methodology , author =. 2022 , pages =. doi:10.1111/rssb.12509 , abstract =

    work page doi:10.1111/rssb.12509 2022

  42. [42]

    IEEE Transactions on Network Science and Engineering , author =

    Clustering of. IEEE Transactions on Network Science and Engineering , author =. 2024 , keywords =. doi:10.1109/TNSE.2024.3374102 , abstract =

    work page doi:10.1109/tnse.2024.3374102 2024

  43. [43]

    Limit results for distributed estimation of invariant subspaces in multiple networks inference and

    Zheng, Runbing and Tang, Minh , month = may, year =. Limit results for distributed estimation of invariant subspaces in multiple networks inference and. doi:10.48550/arXiv.2206.04306 , abstract =

    work page doi:10.48550/arxiv.2206.04306

  44. [44]

    Consistency of community detection in networks under degree-corrected stochastic block models , volume =

    Zhao, Yunpeng and Levina, Elizaveta and Zhu, Ji , month = aug, year =. Consistency of community detection in networks under degree-corrected stochastic block models , volume =. The Annals of Statistics , publisher =. doi:10.1214/12-AOS1036 , abstract =

    work page doi:10.1214/12-aos1036

  45. [45]

    , year =

    Wainwright, Martin J. , year =. High-. doi:10.1017/9781108627771 , abstract =

    work page doi:10.1017/9781108627771

  46. [46]

    Vershynin, Roman , year =. High-. doi:10.1017/9781108231596 , abstract =

    work page doi:10.1017/9781108231596

  47. [47]

    IEEE Transactions on Information Theory , author =

    Matrices. IEEE Transactions on Information Theory , author =. 2024 , keywords =. doi:10.1109/TIT.2023.3331010 , abstract =

    work page doi:10.1109/tit.2023.3331010 2024

  48. [48]

    Journal of Econometrics , author =

    Mixed membership estimation for social networks , volume =. Journal of Econometrics , author =. 2024 , keywords =. doi:10.1016/j.jeconom.2022.12.003 , abstract =

    work page doi:10.1016/j.jeconom.2022.12.003 2024

  49. [49]

    High dimensional deformed rectangular matrices with applications in matrix denoising , volume =

    Ding, Xiucai , month = feb, year =. High dimensional deformed rectangular matrices with applications in matrix denoising , volume =. Bernoulli , publisher =. doi:10.3150/19-BEJ1129 , abstract =

    work page doi:10.3150/19-bej1129

  50. [50]

    The Annals of Statistics , author =

    Chen, Yuxin and Cheng, Chen and Fan, Jianqing , month = feb, year =. Asymmetry helps:. The Annals of Statistics , publisher =. doi:10.1214/20-AOS1963 , abstract =

    work page doi:10.1214/20-aos1963

  51. [51]

    Journal of Machine Learning Research , author =

    Optimal. Journal of Machine Learning Research , author =. 2021 , pages =

    2021

  52. [52]

    Tony and Zhang, Anru , month = feb, year =

    Cai, T. Tony and Zhang, Anru , month = feb, year =. Rate-optimal perturbation bounds for singular subspaces with applications to high-dimensional statistics , volume =. The Annals of Statistics , publisher =. doi:10.1214/17-AOS1541 , abstract =

    work page doi:10.1214/17-aos1541

  53. [53]

    The Annals of Statistics , author =

    Bao, Zhigang and Ding, Xiucai and Wang,. Singular vector and singular subspace distribution for the matrix denoising model , volume =. The Annals of Statistics , publisher =. 2021 , keywords =. doi:10.1214/20-AOS1960 , abstract =

    work page doi:10.1214/20-aos1960 2021

  54. [54]

    IEEE Transactions on Information Theory , author =

    Uncertainty. IEEE Transactions on Information Theory , author =. 2023 , keywords =. doi:10.1109/TIT.2022.3205781 , abstract =

    work page doi:10.1109/tit.2022.3205781 2023

  55. [55]

    Electronic Journal of Statistics , author =

    Xia, Dong , month = jan, year =. Normal approximation and confidence region of singular subspaces , volume =. Electronic Journal of Statistics , publisher =. doi:10.1214/21-EJS1876 , abstract =

    work page doi:10.1214/21-ejs1876

  56. [56]

    Journal of Machine Learning Research , author =

    Robust. Journal of Machine Learning Research , author =. 2024 , pages =

    2024

  57. [57]

    , month = jan, year =

    Tang, Minh and Cape, Joshua R. , month = jan, year =. Eigenvector fluctuations and limit results for random graphs with infinite rank kernels , url =. doi:10.48550/arXiv.2501.15725 , abstract =

    work page doi:10.48550/arxiv.2501.15725

  58. [58]

    IEEE Transactions on Information Theory , author =

    Entrywise. IEEE Transactions on Information Theory , author =. 2022 , keywords =. doi:10.1109/TIT.2022.3159085 , abstract =

    work page doi:10.1109/tit.2022.3159085 2022

  59. [59]

    Estimating shared subspace with

    Yang, Yuepeng and Ma, Cong , month = feb, year =. Estimating shared subspace with. doi:10.48550/arXiv.2501.09336 , abstract =

    work page doi:10.48550/arxiv.2501.09336

  60. [60]

    Bernoulli , author =

    Xie, Fangzheng , month = feb, year =. Entrywise limit theorems for eigenvectors of signal-plus-noise matrix models with weak signals , volume =. Bernoulli , publisher =. doi:10.3150/23-BEJ1602 , abstract =

    work page doi:10.3150/23-bej1602

  61. [61]

    and Zhou, Harrison Y

    Zhang, Anderson Y. and Zhou, Harrison Y. , month = oct, year =. Leave-one-out singular subspace perturbation analysis for spectral clustering , volume =. The Annals of Statistics , publisher =. doi:10.1214/24-AOS2418 , abstract =

    work page doi:10.1214/24-aos2418

  62. [62]

    Computational lower bounds for graphon estimation via low-degree polynomials , volume =

    Luo, Yuetian and Gao, Chao , month = oct, year =. Computational lower bounds for graphon estimation via low-degree polynomials , volume =. The Annals of Statistics , publisher =. doi:10.1214/24-AOS2437 , abstract =

    work page doi:10.1214/24-aos2437

  63. [63]

    Li, Jingyang and Cai, Jian-Feng and Chen, Yang and Xia, Dong , month = oct, year =. Online. doi:10.48550/arXiv.2306.03372 , abstract =

    work page doi:10.48550/arxiv.2306.03372

  64. [64]

    and Bruna, Joan , month = mar, year =

    Damian, Alex and Pillaud-Vivien, Loucas and Lee, Jason D. and Bruna, Joan , month = mar, year =. Computational-. doi:10.48550/arXiv.2403.05529 , abstract =

    work page doi:10.48550/arxiv.2403.05529

  65. [65]

    Tony and Guo, Zijian , month = apr, year =

    Cai, T. Tony and Guo, Zijian , month = apr, year =. Confidence intervals for high-dimensional linear regression:. The Annals of Statistics , publisher =. doi:10.1214/16-AOS1461 , abstract =

    work page doi:10.1214/16-aos1461

  66. [66]

    and Scott, Elizabeth L

    Neyman, J. and Scott, Elizabeth L. , year =. Consistent. Econometrica , publisher =. doi:10.2307/1914288 , number =

    work page doi:10.2307/1914288

  67. [67]

    de la and Montgomery-Smith, S

    Pena, Victor H. de la and Montgomery-Smith, S. J. , month = apr, year =. Decoupling. The Annals of Probability , publisher =. doi:10.1214/aop/1176988291 , abstract =

    work page doi:10.1214/aop/1176988291

  68. [68]

    Tony and Ma, Zongming and Wu, Yihong , month = dec, year =

    Cai, T. Tony and Ma, Zongming and Wu, Yihong , month = dec, year =. Sparse. The Annals of Statistics , publisher =. doi:10.1214/13-AOS1178 , abstract =

    work page doi:10.1214/13-aos1178

  69. [69]

    and Giné, Evarist , editor =

    De La Peña, Víctor H. and Giné, Evarist , editor =. Decoupling , copyright =. 1999 , keywords =. doi:10.1007/978-1-4612-0537-1 , urldate =

    work page doi:10.1007/978-1-4612-0537-1 1999

  70. [70]

    , month = apr, year =

    Ros, Valentina and Fyodorov, Yan V. , month = apr, year =. The high-d landscapes paradigm: spin-glasses, and beyond , shorttitle =. doi:10.48550/arXiv.2209.07975 , abstract =

    work page doi:10.48550/arxiv.2209.07975

  71. [71]

    IEEE Transactions on Information Theory , author =

    Tight. IEEE Transactions on Information Theory , author =. 2011 , keywords =. doi:10.1109/TIT.2011.2111771 , abstract =

    work page doi:10.1109/tit.2011.2111771 2011

  72. [72]

    Probability Theory and Related Fields , author =

    Optimal estimation and rank detection for sparse spiked covariance matrices , volume =. Probability Theory and Related Fields , author =. 2015 , keywords =. doi:10.1007/s00440-014-0562-z , abstract =

    work page doi:10.1007/s00440-014-0562-z 2015

  73. [73]

    Bosch, David and Panahi, Ashkan , month = apr, year =. A. Proceedings of

  74. [74]

    Tony and Low, Mark and Ma, Zongming , month = jul, year =

    Cai, T. Tony and Low, Mark and Ma, Zongming , month = jul, year =. Adaptive. Journal of the American Statistical Association , publisher =. doi:10.1080/01621459.2013.879260 , abstract =

    work page doi:10.1080/01621459.2013.879260 2013

  75. [75]

    Tony and Low, Mark G

    Cai, T. Tony and Low, Mark G. and Xia, Yin , month = apr, year =. Adaptive confidence intervals for regression functions under shape constraints , volume =. The Annals of Statistics , publisher =. doi:10.1214/12-AOS1068 , abstract =

    work page doi:10.1214/12-aos1068

  76. [76]

    Tony and Yuan, Ming , month = sep, year =

    Cai, T. Tony and Yuan, Ming , month = sep, year =. Minimax and. Journal of the American Statistical Association , publisher =. doi:10.1080/01621459.2012.716337 , abstract =

    work page doi:10.1080/01621459.2012.716337 2012

  77. [77]

    Tony and Low, Mark G

    Cai, T. Tony and Low, Mark G. , month = oct, year =. Optimal adaptive estimation of a quadratic functional , volume =. The Annals of Statistics , publisher =. doi:10.1214/009053606000000849 , abstract =

    work page doi:10.1214/009053606000000849

  78. [78]

    Tony and Low, Mark G

    Cai, T. Tony and Low, Mark G. , month = oct, year =. On adaptive estimation of linear functionals , volume =. The Annals of Statistics , publisher =. doi:10.1214/009053605000000633 , abstract =

    work page doi:10.1214/009053605000000633

  79. [79]

    Tony and Low, Mark G

    Cai, T. Tony and Low, Mark G. , month = oct, year =. An adaptation theory for nonparametric confidence intervals , volume =. The Annals of Statistics , publisher =. doi:10.1214/009053604000000049 , abstract =

    work page doi:10.1214/009053604000000049

  80. [80]

    Tony and Low, Mark G

    Cai, T. Tony and Low, Mark G. , month = apr, year =. Minimax estimation of linear functionals over nonconvex parameter spaces , volume =. The Annals of Statistics , publisher =. doi:10.1214/009053604000000094 , abstract =

    work page doi:10.1214/009053604000000094

Showing first 80 references.