Pairwise-Independent Contention Resolution

  • Anupam Gupta
  • , Jinqiao Hu
  • , Gregory Kehne
  • , Roie Levin

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

Abstract

We study online contention resolution schemes (OCRSs) and prophet inequalities for non-product distributions. Specifically, when the active set is sampled according to a pairwise-independent (PI) distribution, we show a (1-ok(1))-selectable OCRS for uniform matroids of rank k, and Ω(1)-selectable OCRSs for laminar, graphic, cographic, transversal, and regular matroids. These imply prophet inequalities with the same ratios when the set of values is drawn according to a PI distribution. Our results complement recent work of Dughmi et al. [14] showing that no ω(1/k)-selectable OCRS exists in the PI setting for general matroids of rank k.

Original languageAmerican English
Title of host publicationInteger Programming and Combinatorial Optimization - 25th International Conference, IPCO 2024, Proceedings
EditorsJens Vygen, Jarosław Byrka
PublisherSpringer Science and Business Media Deutschland GmbH
Pages196-209
Number of pages14
ISBN (Print)9783031598340
DOIs
StatePublished - 2024
Event25th International Conference on Integer Programming and Combinatorial Optimization, IPCO 2024 - Wroclaw, Poland
Duration: Jul 3 2024Jul 5 2024

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume14679 LNCS

Conference

Conference25th International Conference on Integer Programming and Combinatorial Optimization, IPCO 2024
Country/TerritoryPoland
CityWroclaw
Period7/3/247/5/24

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Keywords

  • Contention Resolution
  • Online Algorithms
  • Prophet Inequalities

Fingerprint

Dive into the research topics of 'Pairwise-Independent Contention Resolution'. Together they form a unique fingerprint.

Cite this