English

High-Dimensional Geometric Streaming in Polynomial Space

Data Structures and Algorithms 2022-09-28 v4 Computational Geometry Functional Analysis

Abstract

Many existing algorithms for streaming geometric data analysis have been plagued by exponential dependencies in the space complexity, which are undesirable for processing high-dimensional data sets. In particular, once dlognd\geq\log n, there are no known non-trivial streaming algorithms for problems such as maintaining convex hulls and L\"owner-John ellipsoids of nn points, despite a long line of work in streaming computational geometry since [AHV04]. We simultaneously improve these results to poly(d,logn)\mathrm{poly}(d,\log n) bits of space by trading off with a poly(d,logn)\mathrm{poly}(d,\log n) factor distortion. We achieve these results in a unified manner, by designing the first streaming algorithm for maintaining a coreset for \ell_\infty subspace embeddings with poly(d,logn)\mathrm{poly}(d,\log n) space and poly(d,logn)\mathrm{poly}(d,\log n) distortion. Our algorithm also gives similar guarantees in the \emph{online coreset} model. Along the way, we sharpen results for online numerical linear algebra by replacing a log condition number dependence with a logn\log n dependence, answering a question of [BDM+20]. Our techniques provide a novel connection between leverage scores, a fundamental object in numerical linear algebra, and computational geometry. For p\ell_p subspace embeddings, we give nearly optimal trade-offs between space and distortion for one-pass streaming algorithms. For instance, we give a deterministic coreset using O(d2logn)O(d^2\log n) space and O((dlogn)1/21/p)O((d\log n)^{1/2-1/p}) distortion for p>2p>2, whereas previous deterministic algorithms incurred a poly(n)\mathrm{poly}(n) factor in the space or the distortion [CDW18]. Our techniques have implications in the offline setting, where we give optimal trade-offs between the space complexity and distortion of subspace sketch data structures. To do this, we give an elementary proof of a "change of density" theorem of [LT80] and make it algorithmic.

Keywords

Cite

@article{arxiv.2204.03790,
  title  = {High-Dimensional Geometric Streaming in Polynomial Space},
  author = {David P. Woodruff and Taisuke Yasuda},
  journal= {arXiv preprint arXiv:2204.03790},
  year   = {2022}
}

Comments

Abstract shortened to meet arXiv limits; v2 fix statements concerning online condition number; v3 to appear in FOCS 2022; v4 minor fixes

R2 v1 2026-06-24T10:41:54.791Z