Subexponential time algorithms for finding small tree and path decompositions

6Citations
Citations of this article
3Readers
Mendeley users who have this article in their library.
Get full text

Abstract

The Minimum Size Tree Decomposition (MSTD) and Minimum Size Path Decomposition (MSPD) problems ask for a given nvertex graph G and integer k, what is the minimum number of bags of a tree decomposition (respectively, path decomposition) of width at most k. The problems are known to be NP-complete for each fixed k ≥ 4. In this paper we present algorithms that solve both problems for fixed k in 2O(n/ log n) time and show that they cannot be solved in 2o(n/ log n) time, assuming the Exponential Time Hypothesis.

Cite

CITATION STYLE

APA

Bodlaender, H. L., & Nederlof, J. (2015). Subexponential time algorithms for finding small tree and path decompositions. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9294, pp. 179–190). Springer Verlag. https://doi.org/10.1007/978-3-662-48350-3_16

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