On Labeled Traveling Salesman Problems

  • Authors:
  • Basile Couëtoux;Laurent Gourvès;Jérôme Monnot;Orestis A. Telelis

  • Affiliations:
  • CNRS UMR 7024 LAMSADE, Université de Paris-Dauphine, France;CNRS UMR 7024 LAMSADE, Université de Paris-Dauphine, France;CNRS UMR 7024 LAMSADE, Université de Paris-Dauphine, France;Department of Computer Science, University of Aarhus, Denmark

  • Venue:
  • ISAAC '08 Proceedings of the 19th International Symposium on Algorithms and Computation
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

We consider labeled Traveling Salesman Problems, defined upon acomplete graph of n vertices with colored edges. Theobjective is to find a tour of maximum (or minimum) number ofcolors. We derive results regarding hardness of approximation, andanalyze approximation algorithms for both versions of the problem.For the maximization version we give a $\frac{1}{2}$-approximationalgorithm and show that it is APX-hard. For the minimizationversion, we show that it is not approximable within n1-ε for every ε 0.When every color appears in the graph at most r times andr is an increasing function of n the problem isnot O(r1-ε)-approximable. For fixed constantr we analyze a polynomial-time (r + Hr)/2-approximation algorithm (Hr is the r-th harmonic number), andprove APX-hardness. Analysis of the studied algorithms isshown to be tight.