Approximation algorithms for the maximum leaf spanning tree problem on acyclic digraphs

  • Authors:
  • Nadine Schwartges;Joachim Spoerhase;Alexander Wolff

  • Affiliations:
  • Computer Science I, University of Würzburg, Germany;Computer Science I, University of Würzburg, Germany;Computer Science I, University of Würzburg, Germany

  • Venue:
  • WAOA'11 Proceedings of the 9th international conference on Approximation and Online Algorithms
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

We consider the problem Maximum Leaf Spanning Tree (MLST) on digraphs, which is defined as follows. Given a digraph G, find a directed spanning tree of G that maximizes the number of leaves. MLST is NP-hard. Existing approximation algorithms for MLST have ratios of O(√{OPT}) and 92. We focus on the special case of acyclic digraphs and propose two linear-time approximation algorithms; one with ratio 4 that uses a result of Daligault and Thomassé and one with ratio 2 based on a 3-approximation algorithm of Lu and Ravi for the undirected version of the problem. We complement these positive results by observing that MLST is MaxSNP-hard on acyclic digraphs. Hence, this special case does not admit a PTAS (unless P = NP.