Subcoloring of (Unit) Disk Graphs

TL;DR

This paper studies the subcoloring problem in (unit) disk graphs, providing approximation algorithms, hardness results, and bounds on the subchromatic number, thus advancing understanding of graph coloring generalizations in geometric graphs.

Contribution

It introduces the subcoloring concept for (unit) disk graphs, proves NP-hardness for certain cases, and offers approximation algorithms and bounds on the subchromatic number.

Findings

Every unit disk graph admits a subcoloring with at most 7 colors.

3-Subcoloring can be approximated within a factor of 3 in polynomial time.

NP-hardness results for 2-Subcoloring in specific unit disk graph subclasses.

Abstract

A subcoloring of a graph is a partition of its vertex set into subsets (called colors), each inducing a disjoint union of cliques. It is a natural generalization of the classical proper coloring, in which each color must instead induce an independent set. Similarly to proper coloring, we define the subchromatic number of a graph as the minimum integer k such that it admits a subcoloring with k colors, and the corresponding problem k-Subcoloring which asks whether a graph has subchromatic number at most k. In this paper, we initiate the study of the subcoloring of (unit) disk graphs. One motivation stems from the fact that disk graphs can be seen as a dense generalization of planar graphs where, intuitively, each vertex can be blown into a large clique--much like subcoloring generalizes proper coloring. Interestingly, it can be observed that every unit disk graph admits a subcoloring with at most 7 colors. We first prove that the subchromatic number can be 3-approximated in polynomial-time in unit disk graphs. We then present several hardness results for special cases of unit disk graphs which somehow prevents the use of classical approaches for improving this result. We show in particular that 2-subcoloring remains NP-hard in triangle-free unit disk graphs, as well as in unit disk graphs representable within a strip of bounded height. We also solve an open question of Broersma, Fomin, Ne\v{s}et\v{r}il, and Woeginger (2002) by proving that 3-Subcoloring remains NP-hard in co-comparability graphs. Finally, we prove that every $n$-vertex disk graph admits a subcoloring with at most $O(\log^3(n))$ colors and present a $O(\log^2(n))$-approximation algorithm for computing the subchromatic number of such graphs. This is achieved by defining a decomposition and a special type of co-comparability disk graph, called $\Delta$-disk graphs, which might be of independent interest.