Constant domination-packing ratio per graph class
Question 1.1 · arXiv:2503.05562
Status partial high confidence
Question 1.1 asks for which graph classes the domination-packing ratio $\gamma(G)/\rho(G)$ is bounded by a class-specific constant. The source paper itself gives positive answers for 2-degenerate, AT-free, and unit-disk graphs, with improved bounds for planar graphs and graphs of bounded treewidth or twin-width. A 2026 follow-up (arXiv:2602.18402) extends the positive answers further to chordal bipartite graphs and homogeneously orderable graphs, and provides an improved bound for planar graphs and a direct proof for trees. The general question — whether every natural graph class admits such a constant — remains open.
Cited literature (1)
-
Extends positive answers to the domination-packing ratio question to chordal bipartite graphs and homogeneously orderable graphs, gives an improved bound for planar graphs, and provides a simplified direct proof for trees.
Reviewer notes. The source paper already gives a rich set of positive answers to Question 1.1 for specific graph classes. The follow-up arXiv:2602.18402 (verified via WebFetch) directly continues this research thread. Authors of 2602.18402 were not extracted from the abstract page. The general meta-question of characterising all graph classes with bounded ratio remains open.
Context
The paper studies the ratio between the dominating number $\gamma(G)$ and the packing number $\rho(G)$ of a graph $G$. A positive answer immediately gives the bound $c_{\mathcal{G}}$ on the integrality gap of both the dominating and the packing integer programs for any $G\in\mathcal{G}$.
Source paper
On graph classes with constant domination-packing ratio
Marthe Bonamy, Mónika Csikós, Anna Gujgiczer, Yelena Yuditsky · 2025-03-07
https://arxiv.org/abs/2503.05562