English

Euclidean distortion and the Sparsest Cut

Metric Geometry 2007-05-23 v1

Abstract

We prove that every nn-point metric space of negative type (and, in particular, every nn-point subset of L1L_1) embeds into a Euclidean space with distortion O(lognloglogn)O(\sqrt{\log n} \cdot\log \log n), a result which is tight up to the iterated logarithm factor. As a consequence, we obtain the best known polynomial-time approximation algorithm for the Sparsest Cut problem with general demands. Namely, if the demand is supported on a subset of size kk, we achieve an approximation ratio of O(logkloglogk)O(\sqrt{\log k}\cdot \log \log k).

Keywords

Cite

@article{arxiv.math/0508154,
  title  = {Euclidean distortion and the Sparsest Cut},
  author = {Sanjeev Arora and James R. Lee and Assaf Naor},
  journal= {arXiv preprint arXiv:math/0508154},
  year   = {2007}
}

Comments

20 pages

R2 v1 2026-07-22T17:22:55.135Z