JOHNSON-LINDENSTRAUSS EMBEDDINGS WITH KRONECKER STRUCTURE

Stefan Bamberger, Felix Krahmer, Rachel Ward

Research output: Contribution to journalArticlepeer-review

5 Scopus citations

Abstract

We prove the Johnson-Lindenstrauss property for matrices ΦDξ, where Φ has the restricted isometry property and Dξ is a diagonal matrix containing the entries of a Kronecker product ξ = ξ(1) ⊗ ․․․ ⊗ ξ(d) of d independent Rademacher vectors. Such embeddings have been proposed in recent works for a number of applications concerning compression of tensor structured data, including the oblivious sketching procedure by Ahle et al. for approximate tensor computations. For preserving the norms of p points simultaneously, our result requires Φ to have the restricted isometry property for sparsity C(d)(log p)d. In the case of subsampled Hadamard matrices, this can improve the dependence of the embedding dimension on p to (log p)d while the best previously known result required (log p)d+1. That is, for the case of d = 2 at the core of the oblivious sketching procedure by Ahle et al., the scaling improves from cubic to quadratic. We provide a counterexample to prove that the scaling established in our result is optimal under mild assumptions.

Original languageEnglish
Pages (from-to)1806-1850
Number of pages45
JournalSIAM Journal on Matrix Analysis and Applications
Volume43
Issue number4
DOIs
StatePublished - 2022

Keywords

  • Johnson-Lindenstrauss embeddings
  • Kronecker product
  • higher-order chaos
  • restricted isometry property

Fingerprint

Dive into the research topics of 'JOHNSON-LINDENSTRAUSS EMBEDDINGS WITH KRONECKER STRUCTURE'. Together they form a unique fingerprint.

Cite this