RANDOM SPARSE SAMPLING IN A GIBBS WEIGHTED TREE

Abstract : Let µ be the geometric realization on [0, 1] of a Gibbs measure on Σ = {0, 1} N associated with a Hölder potential. The thermodynamic and multifractal properties of µ are well known to be linked via the multifractal formalism. In this article, the impact of a random sampling procedure on this structure is studied. More precisely, let {Iw}w∈Σ * stand for the collection of dyadic subintervals of [0, 1] naturally indexed by the set of finite dyadic words Σ *. Fix η ∈ (0, 1), and a sequence (pw)w∈Σ * of independent Bernoulli variables of parameters 2 −|w|(1−η) (|w| is the length of w). We consider the (very sparse) remaining values µ = {µ(Iw) : w ∈ Σ * , pw = 1}. We prove that when η < 1/2, it is possible to entirely reconstruct µ from the sole knowledge of µ, while it is not possible when η > 1/2, hence a first phase transition phenomenon. We show that, for all η ∈ (0, 1), it is possible to reconstruct a large part of the initial multifractal structure of µ, via the fine study of µ. After reorganization, these coefficients give rise to a random capacity with new remarkable scaling and multifractal properties: its L q-spectrum exhibits two phase transitions, and has a rich thermodynamic and geometric structure.
Type de document :
Article dans une revue
Journal of the Institute of Mathematics of Jussieu, Cambridge University Press (CUP), A Paraître
Liste complète des métadonnées

Littérature citée [36 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01612286
Contributeur : Stéphane Seuret <>
Soumis le : vendredi 6 octobre 2017 - 16:41:39
Dernière modification le : lundi 4 décembre 2017 - 10:36:47

Fichier

Gibbs_Sampling.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01612286, version 1

Citation

Julien Barral, Stephane Seuret. RANDOM SPARSE SAMPLING IN A GIBBS WEIGHTED TREE. Journal of the Institute of Mathematics of Jussieu, Cambridge University Press (CUP), A Paraître. 〈hal-01612286〉

Partager

Métriques

Consultations de la notice

20

Téléchargements de fichiers

7