arXiv · 2611.12419
Optimal Rates for Adaptive Private $k$-PCA
Abstract
Given $n$ i.i.d. random matrices $A_i \in \mathbb{R}^{d \times d}$ with common expectation $Σ$, the goal of \emph{Differentially Private Stochastic PCA} is to identify a $k$-dimensional subspace capturing the leading variance directions of $Σ$, while preserving differential privacy~(DP) for each individual sample $A_i$. \citet{dungler2025iterative} introduced \kdppca, the first algorithm to simultaneously (1) achieve sample complexity $n=\widetilde O(d)$ for sub-Gaussian data, (2) adapt its privacy noise to the intrinsic randomness of the data, and (3) extend seamlessly to any target dimension $k\le d$. However, its sample complexity has a suboptimal dependence on $k$. We propose the first algorithm that achieves optimal sample complexity in both $d$ and $k$, while retaining properties (2) and (3) of \kdppca. Additionally, prior utility guarantees require a minimum sample size with exponential dependence on the eigengap. Our method removes this dependence. It also improves the dependence on spectral parameters, matching known lower bounds for the spiked covariance model. Unlike deflation-based approaches like \kdppca, our algorithm updates the full $d\times k$ subspace jointly rather than one eigenvector at a time. This non-deflation structure simplifies the algorithm, reduces the number of hyper-parameters, and improves computational efficiency, particularly when using block linear algebra libraries.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Johanna Düngler, Amartya Sanyal. 2026-11-09. Optimal Rates for Adaptive Private $k$-PCA. https://arxiv.science/abs/2611.12419
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite arXiv Science by Indiciae: BibTeX · RIS.