Skip to Main Content


AMS eBook CollectionsOne of the world's most respected mathematical collections, available in digital format for your library or institution


A power law of order 1/4 for critical mean field Swendsen-Wang dynamics

About this Title

Yun Long, Asaf Nachmias, Weiyang Ning and Yuval Peres

Publication: Memoirs of the American Mathematical Society
Publication Year: 2014; Volume 232, Number 1092
ISBNs: 978-1-4704-0910-4 (print); 978-1-4704-1895-3 (online)
DOI: https://doi.org/10.1090/memo/1092
Published electronically: March 11, 2014
Keywords: Markov chains, Mixing time, Ising model, Swendsen-Wang algorithm
MSC: Primary 60J10, 82B20

View full volume PDF

View other years and numbers:

Table of Contents

Chapters

  • 1. Introduction
  • 2. Statement of the results
  • 3. Mixing time preliminaries
  • 4. Outline of the proof of Theorem
  • 5. Random graph estimates
  • 6. Supercritical case
  • 7. Subcritical case
  • 8. Critical Case
  • 9. Fast mixing of the Swendsen-Wang process on trees
  • Acknowledgements

Abstract

The Swendsen-Wang dynamics is a Markov chain widely used by physicists to sample from the Boltzmann-Gibbs distribution of the Ising model. Cooper, Dyer, Frieze and Rue proved that on the complete graph $K_n$ the mixing time of the chain is at most $O(\sqrt {n})$ for all non-critical temperatures. In this paper we show that the mixing time is $\Theta (1)$ in high temperatures, $\Theta (\log n)$ in low temperatures and $\Theta (n^{1/4})$ at criticality. We also provide an upper bound of $O(\log n)$ for Swendsen-Wang dynamics for the $q$-state ferromagnetic Potts model on any tree of $n$ vertices.

References [Enhancements On Off] (What's this?)

References