Inference and Uncertainty Quantification for Streaming r-PCA

Abstract

We address two open questions in streaming PCA via Oja's algorithm: sharp operator-norm convergence for general rank under sub-Gaussian data, and distributional inference for the resulting subspace estimator. Existing convergence analyses, even in the rank-one case, either assume bounded data or leave non-vanishing remainder terms that prevent adaptation to a polynomially vanishing tail spectrum, while existing distributional results are confined to the rank-one case. Our convergence theory removes these remainder terms and yields a sharp rate. In the dense-tail spiked covariance regime, this rate matches the minimax rate up to logarithmic factors. More generally, we prove a matching lower bound, up to logarithmic factors, across both dense-tail and sparse-tail regimes under a mild nondegeneracy condition. The analysis yields a linearization of Oja's iterates, which in turn enables a high-dimensional Gaussian approximation for the general-rank subspace estimation error with an explicit limiting covariance. We also establish a row-wise Gaussian approximation over convex sets for the aligned difference, recovering prior rank-one results as special cases. For practical inference, we develop an online multiplier bootstrap algorithm and prove its consistency. Beyond streaming PCA, our techniques contribute to Gaussian approximation and bootstrap inference for nonconvex stochastic approximation.

Improving ecological inference and uncertainty quantification from camera trap data through the fusion of AI confidences and manual annotations
Adira Cohen, Erin M. Schliep, Mohammad Alyetama +2
arXiv14 Oct 2026PreprintarXiv licenceFull textStatisticsarXiv
Uncertainty Quantification of State Variables Trajectories in the Context of Inverse Problems: An Approach from Bayesian Inference and FDA
Luis Alejandro Baena-Marín, Juan Daniel Molina, Juan Camilo Bermúdez-Colorado +1
arXiv1 Sep 2026PreprintarXiv licenceFull textStatisticsarXiv
MCMC-Free Uncertainty Quantification for Deep Generative Inference for Spatially Varying Coefficient Models at Scale
Yeseul Jeon, Aaron Scheffler, Rajarshi Guhaniyogi
arXiv28 Sep 2026PreprintarXiv licenceFull textStatisticsarXiv
Bayesian Uncertainty Quantification for fMRI Functional Connectivity via Simulation-Based Inference
Simon Carter, Zeming Kuang, Lilianne R. Mujica-Parodi +1
arXiv24 Sep 2026PreprintarXiv licenceFull textStatisticsarXiv
Global Parameter Identification and Uncertainty Quantification of reduced-order ADM1 using Maximum Likelihood Estimation and Bayesian Inference
Simon Hellmann, Leander Lerch, Terrance Wilms +6
arXiv29 Oct 2026PreprintCC BY-NC-ND 4.0Full textStatisticsarXiv

Cite

BibTeX RIS

BibTeX

@misc{xu2026inference,
  title = {{Inference and Uncertainty Quantification for Streaming $r$-PCA}},
  author = {Haoshu Xu and Hongzhe Li},
  howpublished = {arXiv},
  year = {2026},
  month = nov,
  doi = {10.48550/arxiv.2608.18374},
  eprint = {2608.18374},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2608.18374},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Inference and Uncertainty Quantification for Streaming r-PCA
AU  - Haoshu Xu
AU  - Hongzhe Li
PY  - 2026
DA  - 2026/11/04
PB  - arXiv
DO  - 10.48550/arxiv.2608.18374
UR  - https://arxiv.org/abs/2608.18374
AB  - We address two open questions in streaming PCA via Oja's algorithm: sharp operator-norm convergence for general rank under sub-Gaussian data, and distributional inference for the resulting subspace estimator. Existing convergence analyses, even in the rank-one case, either assume bounded data or leave non-vanishing remainder terms that prevent adaptation to a polynomially vanishing tail spectrum, while existing distributional results are confined to the rank-one case. Our convergence theory removes these remainder terms and yields a sharp rate. In the dense-tail spiked covariance regime, this rate matches the minimax rate up to logarithmic factors. More generally, we prove a matching lower bound, up to logarithmic factors, across both dense-tail and sparse-tail regimes under a mild nondegeneracy condition. The analysis yields a linearization of Oja's iterates, which in turn enables a high-dimensional Gaussian approximation for the general-rank subspace estimation error with an explicit limiting covariance. We also establish a row-wise Gaussian approximation over convex sets for the aligned difference, recovering prior rank-one results as special cases. For practical inference, we develop an online multiplier bootstrap algorithm and prove its consistency. Beyond streaming PCA, our techniques contribute to Gaussian approximation and bootstrap inference for nonconvex stochastic approximation.
ER  -

Abstract

Related

Cite