Generalizing the Johnson-Lindenstrauss lemma to k-dimensional affine subspaces

Alon Dmitriyuk; Yehoram Gordon

Studia Mathematica (2009)

  • Volume: 195, Issue: 3, page 227-241
  • ISSN: 0039-3223

Abstract

top
Let ε > 0 and 1 ≤ k ≤ n and let W l l = 1 p be affine subspaces of ℝⁿ, each of dimension at most k. Let m = O ( ε - 2 ( k + l o g p ) ) if ε < 1, and m = O(k + log p/log(1 + ε)) if ε ≥ 1. We prove that there is a linear map H : m such that for all 1 ≤ l ≤ p and x , y W l we have ||x-y||₂ ≤ ||H(x)-H(y)||₂ ≤ (1+ε)||x-y||₂, i.e. the distance distortion is at most 1 + ε. The estimate on m is tight in terms of k and p whenever ε < 1, and is tight on ε,k,p whenever ε ≥ 1. We extend these results to embeddings into general normed spaces Y.

How to cite

top

Alon Dmitriyuk, and Yehoram Gordon. "Generalizing the Johnson-Lindenstrauss lemma to k-dimensional affine subspaces." Studia Mathematica 195.3 (2009): 227-241. <http://eudml.org/doc/285147>.

@article{AlonDmitriyuk2009,
abstract = {Let ε > 0 and 1 ≤ k ≤ n and let $\{W_\{l\}\}_\{l=1\}^\{p\}$ be affine subspaces of ℝⁿ, each of dimension at most k. Let $m = O(ε^\{-2\}(k + log p))$ if ε < 1, and m = O(k + log p/log(1 + ε)) if ε ≥ 1. We prove that there is a linear map $H: ℝⁿ → ℝ^\{m\}$ such that for all 1 ≤ l ≤ p and $x,y ∈ W_\{l\}$ we have ||x-y||₂ ≤ ||H(x)-H(y)||₂ ≤ (1+ε)||x-y||₂, i.e. the distance distortion is at most 1 + ε. The estimate on m is tight in terms of k and p whenever ε < 1, and is tight on ε,k,p whenever ε ≥ 1. We extend these results to embeddings into general normed spaces Y.},
author = {Alon Dmitriyuk, Yehoram Gordon},
journal = {Studia Mathematica},
keywords = {local theory; Gaussian processes; high dimensional geometry; convexity; normed linear spaces; Gaussian operators},
language = {eng},
number = {3},
pages = {227-241},
title = {Generalizing the Johnson-Lindenstrauss lemma to k-dimensional affine subspaces},
url = {http://eudml.org/doc/285147},
volume = {195},
year = {2009},
}

TY - JOUR
AU - Alon Dmitriyuk
AU - Yehoram Gordon
TI - Generalizing the Johnson-Lindenstrauss lemma to k-dimensional affine subspaces
JO - Studia Mathematica
PY - 2009
VL - 195
IS - 3
SP - 227
EP - 241
AB - Let ε > 0 and 1 ≤ k ≤ n and let ${W_{l}}_{l=1}^{p}$ be affine subspaces of ℝⁿ, each of dimension at most k. Let $m = O(ε^{-2}(k + log p))$ if ε < 1, and m = O(k + log p/log(1 + ε)) if ε ≥ 1. We prove that there is a linear map $H: ℝⁿ → ℝ^{m}$ such that for all 1 ≤ l ≤ p and $x,y ∈ W_{l}$ we have ||x-y||₂ ≤ ||H(x)-H(y)||₂ ≤ (1+ε)||x-y||₂, i.e. the distance distortion is at most 1 + ε. The estimate on m is tight in terms of k and p whenever ε < 1, and is tight on ε,k,p whenever ε ≥ 1. We extend these results to embeddings into general normed spaces Y.
LA - eng
KW - local theory; Gaussian processes; high dimensional geometry; convexity; normed linear spaces; Gaussian operators
UR - http://eudml.org/doc/285147
ER -

NotesEmbed ?

top

You must be logged in to post comments.

To embed these notes on your page include the following JavaScript code on your page where you want the notes to appear.

Only the controls for the widget will be shown in your chosen language. Notes will be shown in their authored language.

Tells the widget how many notes to show per page. You can cycle through additional notes using the next and previous controls.

    
                

Note: Best practice suggests putting the JavaScript code just before the closing </body> tag.