What can be approximated locally? Case study : Dominating sets in planar graphs

Abstract

Whether local algorithms can compute constant approximations of NP-hard problems is of both practical and theoretical interest. So far, no algorithms achieving this goal are known, as either the approximation ratio or the running time exceed O(l), or the nodes are provided with non-trivial additional information. In this paper, we present the first distributed algorithm approximating a minimum dominating set on a planar graph within a constant factor in constant time. Moreover, the nodes do not need any additional information.