License: Creative Commons Attribution 4.0 International license (CC BY 4.0)
When quoting this document, please refer to the following
DOI: 10.4230/LIPIcs.ISAAC.2021.4
URN: urn:nbn:de:0030-drops-154377
URL: http://dagstuhl.sunsite.rwth-aachen.de/volltexte/2021/15437/
Go to the corresponding LIPIcs Volume Portal

Matheny, Michael ; Phillips, Jeff M.

Approximate Maximum Halfspace Discrepancy

pdf-format:
LIPIcs-ISAAC-2021-4.pdf (1 MB)

Abstract

Consider the geometric range space (X, H_d) where X ⊂ ℝ^d and H_d is the set of ranges defined by d-dimensional halfspaces. In this setting we consider that X is the disjoint union of a red and blue set. For each halfspace h ∈ H_d define a function Φ(h) that measures the "difference" between the fraction of red and fraction of blue points which fall in the range h. In this context the maximum discrepancy problem is to find the h^* = arg max_{h ∈ (X, H_d)} Φ(h). We aim to instead find an ĥ such that Φ(h^*) - Φ(ĥ) ≤ ε. This is the central problem in linear classification for machine learning, in spatial scan statistics for spatial anomaly detection, and shows up in many other areas. We provide a solution for this problem in O(|X| + (1/ε^d) log⁴ (1/ε)) time, for constant d, which improves polynomially over the previous best solutions. For d = 2 we show that this is nearly tight through conditional lower bounds. For different classes of Φ we can either provide a Ω(|X|^{3/2 - o(1)}) time lower bound for the exact solution with a reduction to APSP, or an Ω(|X| + 1/ε^{2-o(1)}) lower bound for the approximate solution with a reduction to 3Sum.
A key technical result is a ε-approximate halfspace range counting data structure of size O(1/ε^d) with O(log (1/ε)) query time, which we can build in O(|X| + (1/ε^d) log⁴ (1/ε)) time.

BibTeX - Entry

@InProceedings{matheny_et_al:LIPIcs.ISAAC.2021.4,
  author =	{Matheny, Michael and Phillips, Jeff M.},
  title =	{{Approximate Maximum Halfspace Discrepancy}},
  booktitle =	{32nd International Symposium on Algorithms and Computation (ISAAC 2021)},
  pages =	{4:1--4:15},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-214-3},
  ISSN =	{1868-8969},
  year =	{2021},
  volume =	{212},
  editor =	{Ahn, Hee-Kap and Sadakane, Kunihiko},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/opus/volltexte/2021/15437},
  URN =		{urn:nbn:de:0030-drops-154377},
  doi =		{10.4230/LIPIcs.ISAAC.2021.4},
  annote =	{Keywords: range spaces, halfspaces, scan statistics, fine-grained complexity}
}

Keywords: range spaces, halfspaces, scan statistics, fine-grained complexity
Collection: 32nd International Symposium on Algorithms and Computation (ISAAC 2021)
Issue Date: 2021
Date of publication: 30.11.2021

DROPS-Home | Fulltext Search | Imprint | Privacy Published by LZI