Small h-coloring problems for bounded degree digraphs

1Citations
Citations of this article
1Readers
Mendeley users who have this article in their library.
Get full text

Abstract

An NP-complete coloring or homomorphism problem may become polynomial time solvable when restricted to graphs with degrees bounded by a small number, but remain NP-complete if the bound is higher. For instance, 3-colorability of graphs with degrees bounded by 3 can be decided by Brooks' theorem, while for graphs with degrees bounded by 4, the 3-colorability problem is NP-complete. We investigate an analogous phenomenon for digraphs, focusing on the three smallest digraphs H with NP-complete H-colorability problems. It turns out that in all three cases the H-coloring problem is polynomial time solvable for digraphs with in-degrees at most 1, regardless of the out-degree bound (respectively with out-degrees at most 1, regardless of the in-degree bound). On the other hand, as soon as both in-and out-degrees are bounded by constants greater than or equal to 2, all three problems are again NP-complete. A conjecture proposed for graphs H by Feder, Hell and Huang states that any variant of the H-coloring problem which is NP-complete without degree constraints is also NP-complete with degree constraints, provided the degree bounds are high enough. Thus, our results verify the conjecture with very precise bounds on both in-and out-degrees that are needed for NP-completeness; in particular, the bounds underscore the fact that the sufficiently large bound must apply to both the in-degrees and the out-degrees. © 2013 Springer-Verlag Berlin Heidelberg.

Cite

CITATION STYLE

APA

Hell, P., & Mishra, A. (2013). Small h-coloring problems for bounded degree digraphs. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7936 LNCS, pp. 579–590). https://doi.org/10.1007/978-3-642-38768-5_51

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free