Open Access
October 2008 Near-minimal spanning trees: A scaling exponent in probability models
David J. Aldous, Charles Bordenave, Marc Lelarge
Ann. Inst. H. Poincaré Probab. Statist. 44(5): 962-976 (October 2008). DOI: 10.1214/07-AIHP138

Abstract

We study the relation between the minimal spanning tree (MST) on many random points and the “near-minimal” tree which is optimal subject to the constraint that a proportion δ of its edges must be different from those of the MST. Heuristics suggest that, regardless of details of the probability model, the ratio of lengths should scale as 1+Θ(δ2). We prove this scaling result in the model of the lattice with random edge-lengths and in the Euclidean model.

Nous étudions la relation entre l’arbre couvrant minimal (ACM) sur des points aléatoires et l’arbre “quasi” optimal sous la contrainte qu’une proportion δ de ses arêtes soit différente de celles de l’ACM. Un raisonnement heuristique suggère que quelque soit le modèle probabiliste sous-jacent, le ratio des longueurs des deux arbres doit varier en 1+Θ(δ2). Nous montrons ce résultat d’échelle pour le modèle de la grille avec des longueurs d’arêtes aléatoires et pour le modèle Euclidien.

Citation

Download Citation

David J. Aldous. Charles Bordenave. Marc Lelarge. "Near-minimal spanning trees: A scaling exponent in probability models." Ann. Inst. H. Poincaré Probab. Statist. 44 (5) 962 - 976, October 2008. https://doi.org/10.1214/07-AIHP138

Information

Published: October 2008
First available in Project Euclid: 24 September 2008

zbMATH: 1186.05108
MathSciNet: MR2453778
Digital Object Identifier: 10.1214/07-AIHP138

Subjects:
Primary: 05C80 , 60K35 , 68W40

Keywords: Combinatorial optimization , continuum percolation , Disordered lattice , Local weak convergence , Minimal spanning tree , Poisson point process , Probabilistic analysis of algorithms , Random geometric graph

Rights: Copyright © 2008 Institut Henri Poincaré

Vol.44 • No. 5 • October 2008
Back to Top