Removing the log Factor from (min, +)-Products on Bounded Range Integer Matrices

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We revisit the problem of multiplying two square matrices over the (min, +) semi-ring, where all entries are integers from a bounded range [−M: M] ∪ {∞}. The current state of the art for this problem is a simple O(Mnω log M) time algorithm by Alon, Galil and Margalit [JCSS'97], where ω is the exponent in the runtime of the fastest matrix multiplication (FMM) algorithm. We design a new simple algorithm whose runtime is O(Mnω + Mn2 log M), thereby removing the log M factor in the runtime if ω > 2 or if nω = Ω(n2 log n).

Original languageEnglish
Title of host publication32nd Annual European Symposium on Algorithms, ESA 2024
EditorsTimothy Chan, Johannes Fischer, John Iacono, Grzegorz Herman
PublisherSchloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing
ISBN (Electronic)9783959773386
DOIs
StatePublished - Sep 2024
Event32nd Annual European Symposium on Algorithms, ESA 2024 - London, United Kingdom
Duration: 2 Sep 20244 Sep 2024

Publication series

NameLeibniz International Proceedings in Informatics, LIPIcs
Volume308

Conference

Conference32nd Annual European Symposium on Algorithms, ESA 2024
Country/TerritoryUnited Kingdom
CityLondon
Period2/09/244/09/24

Keywords

  • (min, +)-product
  • FFT
  • FMM

All Science Journal Classification (ASJC) codes

  • Software

Cite this