STABLE SHREDDED SPHERES AND CAUSAL RANDOM MAPS WITH LARGE FACES

Jakob Björnberg, Nicolas Curien, Sigurdur Örn Stefánsson

Research output: Contribution to journalArticlepeer-review

Abstract

We introduce a new familiy of random compact metric spaces Sα for α ∈ (1, 2), which we call stable shredded spheres. They are constructed from excursions of α-stable Lévy processes on [0, 1] possessing no negative jumps. Informally, viewing the graph of the Lévy excursion in the plane, each jump of the process is “cut open” and replaced by a circle, and then all points on the graph at equal height, which are not separated by a jump, are identified. We show that the shredded spheres arise as scaling limits of models of causal random planar maps with large faces introduced by Di Francesco and Guitter. We also establish that their Hausdorff dimension is almost surely equal to α. Point identification in the shredded spheres is intimately connected to the presence of decrease points in stable spectrally positive Lévy processes, as studied by Bertoin in the 1990s.

Original languageEnglish
Pages (from-to)2056-2084
Number of pages29
JournalAnnals of Probability
Volume50
Issue number5
DOIs
Publication statusPublished - Sep 2022

Bibliographical note

Funding Information:
Funding. The first author acknowledges support from Vetenskapsrådet, Grants 2015-0519 and 2019-04185 and is grateful for the hospitality at Université Paris-Sud Orsay. The second author acknowledges supports from ERC “GeoBrown” as well as the grant ANR-14-CE25-0014 “ANR GRAAL”. The third author acknowledges support from the Icelandic Research Fund, Grant Number: 185233-051, and is grateful for the hospitality at Université Paris-Sud Orsay and at Chalmers.

Publisher Copyright:
© Institute of Mathematical Statistics, 2022

Other keywords

  • Gromov–hausdorff convergence
  • Hausdorff dimension
  • Random planar map
  • Scaling limit
  • Stable distribution

Fingerprint

Dive into the research topics of 'STABLE SHREDDED SPHERES AND CAUSAL RANDOM MAPS WITH LARGE FACES'. Together they form a unique fingerprint.

Cite this