Random 2-cell embeddings of multistars

03/08/2021 ∙ by Jesse Campion Loth, et al. ∙ 0

By using permutation representations of maps, one obtains a bijection between all maps whose underlying graph is isomorphic to a graph G and products of permutations of given cycle types. By using statistics on cycle distributions in products of permutations, one can derive information on the set of all 2-cell embeddings of G. In this paper, we study multistars – loopless multigraphs in which there is a vertex incident with all the edges. The well known genus distribution of the two-vertex multistar, also known as a dipole, can be used to determine the expected genus of the dipole. We then use a result of Stanley to show that, in general, the expected genus of every multistar with n nonleaf edges lies in an interval of length 2/(n+1) centered at the expected genus of an n-edge dipole. As an application, we show that the face distribution of the multistar is the same as the face distribution gained when adding a new vertex to a 2-cell embedded graph, and use this to obtain a general upper bound for the expected number of faces in random embeddings of graphs.



There are no comments yet.


page 1

page 2

page 3

page 4

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.