Creative Commons Attribution 3.0 Unported license
We show that counting Euler tours in undirected bounded tree-width graphs is tractable even in parallel - by proving a GapL upper bound. This is in stark contrast to #P-completeness of the same problem in general graphs. Our main technical contribution is to show how (an instance of) dynamic programming on bounded clique-width graphs can be performed efficiently in parallel. Thus we show that the sequential result of Espelage, Gurski and Wanke for efficiently computing Hamiltonian paths in bounded clique-width graphs can be adapted in the parallel setting to count the number of Hamiltonian paths which in turn is a tool for counting the number of Euler tours in bounded tree-width graphs. Our technique also yields parallel algorithms for counting longest paths and bipartite perfect matchings in bounded-clique width graphs. While establishing that counting Euler tours in bounded tree-width graphs can be computed by non-uniform monotone arithmetic circuits of polynomial degree (which characterize #SAC^1) is relatively easy, establishing a uniform #SAC^1 bound needs a careful use of polynomial interpolation.
@InProceedings{balaji_et_al:LIPIcs.FSTTCS.2015.246,
author = {Balaji, Nikhil and Datta, Samir and Ganesan, Venkatesh},
title = {{Counting Euler Tours in Undirected Bounded Treewidth Graphs}},
booktitle = {35th IARCS Annual Conference on Foundations of Software Technology and Theoretical Computer Science (FSTTCS 2015)},
pages = {246--260},
series = {Leibniz International Proceedings in Informatics (LIPIcs)},
ISBN = {978-3-939897-97-2},
ISSN = {1868-8969},
year = {2015},
volume = {45},
editor = {Harsha, Prahladh and Ramalingam, G.},
publisher = {Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
address = {Dagstuhl, Germany},
URL = {https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.FSTTCS.2015.246},
URN = {urn:nbn:de:0030-drops-56493},
doi = {10.4230/LIPIcs.FSTTCS.2015.246},
annote = {Keywords: Euler Tours, Bounded Treewidth, Bounded clique-width, Hamiltonian cycles, Parallel algorithms}
}