Skip to main navigation Skip to search Skip to main content

Robust Bichromatic Classification Using Two Lines

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

107 Downloads (Pure)

Abstract

Given two sets R and B of n points in the plane, we present efficient algorithms to find a two-line linear classifier that best separates the "red" points in R from the "blue" points in B and is robust to outliers. More precisely, we find a region 𝒲_B bounded by two lines, so either a halfplane, strip, wedge, or double wedge, containing (most of) the blue points B, and few red points. Our running times vary between optimal O(nlog n) up to around O(n³), depending on the type of region 𝒲_B and whether we wish to minimize only red outliers, only blue outliers, or both.
Original languageEnglish
Title of host publication35th International Symposium on Algorithms and Computation (ISAAC 2024)
Editors Julián Mestre, Anthony Wirth
PublisherSchloss Dagstuhl - Leibniz-Zentrum für Informatik
Pages33:1-33:14
Number of pages14
ISBN (Electronic)978-3-95977-354-6
DOIs
Publication statusPublished - 4 Dec 2024
Event35th International Symposium on Algorithms and Computation, ISAAC 2024 - Sydney, Australia
Duration: 8 Dec 2024 → 11 Dec 2024

Publication series

NameLeibniz International Proceedings in Informatics (LIPIcs)
Volume322
ISSN (Print)1868-8969

Conference

Conference35th International Symposium on Algorithms and Computation, ISAAC 2024
Abbreviated titleISAAC 2024
Country/TerritoryAustralia
CitySydney
Period8/12/24 → 11/12/24

Keywords

  • Bichromatic
  • Classification
  • Duality
  • Geometric Algorithms
  • Separating Line

Fingerprint

Dive into the research topics of 'Robust Bichromatic Classification Using Two Lines'. Together they form a unique fingerprint.

Cite this