Improved algorithms for path, matching, and packing problems

  • Authors:

  • Jianer Chen;Songjian Lu;Sing-Hoi Sze;Fenghui Zhang

  • Affiliations:

  • Texas A&M University, College Station, TX;Texas A&M University, College Station, TX;Texas A&M University, College Station, TX;Texas A&M University, College Station, TX

  • Venue:
  • SODA '07 Proceedings of the eighteenth annual ACM-SIAM symposium on Discrete algorithms
  • Year:
  • 2007

Quantified Score

Hi-index 0.01

Visualization

Abstract

Improved randomized and deterministic algorithms are presented for PATH, MATCHING, and PACKING problems. Our randomized algorithms are based on the divide-and-conquer technique, and improve previous best algorithms for these problems. For example, for the k-PATH problem, our randomized algorithm runs in time O(4kk3.42m) and space O(nklogk + m), improving the previous best randomized algorithm for the problem that runs in time O(5.44kkm) and space O(2kkn + m). To achieve improved deterministic algorithms, we study a number of previously proposed de-randomization schemes, and also develop a new derandomization scheme. These studies result in a number of deterministic algorithms: one of time O(4k+o(k)m) for the k-PATH problem, one of time O(2.803kk nlog2 n) for the 3-D MATCHING problem, and one of time O(43k+o(k)n) for the 3-SET PACKING problem. All these significantly improve previous best algorithms for the problems.