Arrow Research search
Back to I&C

I&C 2018

An improved combinatorial algorithm for Boolean matrix multiplication

Journal Article journal-article Computer Science · Theoretical Computer Science

Abstract

We present a new combinatorial algorithm for triangle finding and Boolean matrix multiplication that runs in O ˆ ( n 3 / log 4 ⁡ n ) time, where the O ˆ notation suppresses poly(loglog) factors. This improves the previous best combinatorial algorithm by Chan that runs in O ˆ ( n 3 / log 3 ⁡ n ) time. Our algorithm generalizes the divide-and-conquer strategy of Chan's algorithm. Moreover, we propose a general framework for detecting triangles in graphs and computing Boolean matrix multiplication. Roughly speaking, if we can find the “easy parts” of a given instance efficiently, we can solve the whole problem faster than n 3.

Authors

Keywords

  • Boolean matrix multiplication
  • Combinatorial algorithm
  • Recursion

Context

Venue
Information and Computation
Archive span
1987-2026
Indexed papers
3021
Paper id
13639121144456634
v2026.09.13