English

Parameterized Algorithms for Graph Partitioning Problems

Data Structures and Algorithms 2014-03-04 v1

Abstract

We study a broad class of graph partitioning problems, where each problem is specified by a graph G=(V,E)G=(V,E), and parameters kk and pp. We seek a subset UVU\subseteq V of size kk, such that α1m1+α2m2\alpha_1m_1 + \alpha_2m_2 is at most (or at least) pp, where α1,α2R\alpha_1,\alpha_2\in\mathbb{R} are constants defining the problem, and m1,m2m_1, m_2 are the cardinalities of the edge sets having both endpoints, and exactly one endpoint, in UU, respectively. This class of fixed cardinality graph partitioning problems (FGPP) encompasses Max (k,nk)(k,n-k)-Cut, Min kk-Vertex Cover, kk-Densest Subgraph, and kk-Sparsest Subgraph. Our main result is an O(4k+o(k)Δk)O^*(4^{k+o(k)}\Delta^k) algorithm for any problem in this class, where Δ1\Delta \geq 1 is the maximum degree in the input graph. This resolves an open question posed by Bonnet et al. [IPEC 2013]. We obtain faster algorithms for certain subclasses of FGPPs, parameterized by pp, or by (k+p)(k+p). In particular, we give an O(4p+o(p))O^*(4^{p+o(p)}) time algorithm for Max (k,nk)(k,n-k)-Cut, thus improving significantly the best known O(pp)O^*(p^p) time algorithm.

Keywords

Cite

@article{arxiv.1403.0099,
  title  = {Parameterized Algorithms for Graph Partitioning Problems},
  author = {Hadas Shachnai and Meirav Zehavi},
  journal= {arXiv preprint arXiv:1403.0099},
  year   = {2014}
}
R2 v1 2026-06-22T03:18:20.969Z