Successive minimum spanning trees
From MaRDI portal
Publication:6319955
DOI10.4230/LIPICS.APPROX-RANDOM.2019.60arXiv1906.01533MaRDI QIDQ6319955FDOQ6319955
Authors: Svante Janson, Gregory B. Sorkin
Publication date: 4 June 2019
Abstract: In a complete graph with edge weights drawn independently from a uniform distribution (or alternatively an exponential distribution ), let be the MST (the spanning tree of minimum weight) and let be the MST after deletion of the edges of all previous trees , . We show that each tree's weight converges in probability to a constant with , and we conjecture that . The problem is distinct from that of Frieze and Johansson (2018), finding MSTs of combined minimum weight, and for ours has strictly larger cost. Our results also hold (and mostly are derived) in a multigraph model where edge weights for each vertex pair follow a Poisson process; here we additionally have . Thinking of an edge of weight as arriving at time , Kruskal's algorithm defines forests , each initially empty and eventually equal to , with each arriving edge added to the first where it does not create a cycle. Using tools of inhomogeneous random graphs we obtain structural results including that , the fraction of vertices in the largest component of , converges in probability to a function , uniformly for all , and that a giant component appears in at a time . We conjecture that the functions tend to time translations of a single function, as , uniformly in . Simulations and numerical computations give estimated values of for small , and support the conjectures just stated.
This page was built for publication: Successive minimum spanning trees
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6319955)