Applications of Chebyshev polynomials to low-dimensional computational geometry

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We apply the polynomial method - specifically, Chebyshev polynomials - to obtain a number of new results on geometric approximation algorithms in low constant dimensions. For example, we give an algorithm for constructing ϵ-kernels (coresets for approximate width and approximate convex hull) in close to optimal time O(n + (1/ϵ)(d-1)/2), up to a small near-(1/ϵ)3/2/factor, for any d-dimensional n-point set. We obtain an improved data structure for Euclidean approximate nearest neighbor search with close to O(n log n + (1/ϵ)d/4n) preprocessing time and O((1/ϵ)d/4 log n) query time. We obtain improved approximation algorithms for discrete Voronoi diagrams, diameter, and bichromatic closest pair in the Ls-metric for any even integer constant s ≥ 2. The techniques are general and may have further applications.

Original languageEnglish (US)
Title of host publication33rd International Symposium on Computational Geometry, SoCG 2017
EditorsMatthew J. Katz, Boris Aronov
PublisherSchloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing
Pages261-2615
Number of pages2355
ISBN (Electronic)9783959770385
DOIs
StatePublished - Jun 1 2017
Event33rd International Symposium on Computational Geometry, SoCG 2017 - Brisbane, Australia
Duration: Jul 4 2017Jul 7 2017

Publication series

NameLeibniz International Proceedings in Informatics, LIPIcs
Volume77
ISSN (Print)1868-8969

Other

Other33rd International Symposium on Computational Geometry, SoCG 2017
Country/TerritoryAustralia
CityBrisbane
Period7/4/177/7/17

Keywords

  • Approximate nearest neighbor search
  • Coresets
  • Diameter
  • Streaming
  • The polynomial method

ASJC Scopus subject areas

  • Software

Fingerprint

Dive into the research topics of 'Applications of Chebyshev polynomials to low-dimensional computational geometry'. Together they form a unique fingerprint.

Cite this