Chanal Krauth 2010

From Werner KRAUTH

(Difference between revisions)
Jump to: navigation, search
Revision as of 23:35, 28 January 2011
Werner (Talk | contribs)

← Previous diff
Revision as of 12:02, 29 January 2011
Werner (Talk | contribs)

Next diff →
Line 1: Line 1:
-'''C. Chanal, W. Krauth [http://arxiv.org/pdf/0910.1530v1 ''Convergence and coupling for spin glasses and hard spheres''] Physical Review E ''' 81''' 016705 (2010)'''+'''C. Chanal, W. Krauth Physical Review E ''' 81''' 016705 (2010)'''
 +'''Abstract: ''' We discuss convergence and coupling of Markov chains, and present general
 +relations between the transfer matrices describing these two processes.
 +We then analyze a recently developed local-patch algorithm, which
 +computes rigorous upper bound for the coupling time of a Markov chain for
 +non-trivial statistical-mechanics models. Using the "coupling from
 +the past" protocol, this allows one to exactly sample the underlying
 +equilibrium distribution. For spin glasses in two and three spatial
 +dimensions, the local-patch algorithm works at lower temperatures than
 +previous exact-sampling methods. We discuss variants of the algorithm
 +which might allow one to reach, in three dimensions, the spin-glass
 +transition temperature. The algorithm can be adapted to hard-sphere
 +models. For two-dimensional hard disks, the algorithm allows us to draw
 +exact samples at higher densities than previously possible.
 +[http://arxiv.org/pdf/0910.1530v1 Electronic version (arXiv)]
-Long version of the Chanal Krauth (2008) paper, with +'''Comment''': Long version of the Chanal Krauth (2008) paper, containing an extension of the patch algorithm to hard spheres.
-[http://www.phys.ens.fr/~krauth/get.php?fichier=pruning_ND.py Python implementation] of the patch algorithm+ 
 +[http://www.phys.ens.fr/~krauth/get.php?fichier=pruning_ND.py Python implementation of the patch algorithm]

Revision as of 12:02, 29 January 2011

C. Chanal, W. Krauth Physical Review E 81 016705 (2010)

Abstract: We discuss convergence and coupling of Markov chains, and present general relations between the transfer matrices describing these two processes. We then analyze a recently developed local-patch algorithm, which computes rigorous upper bound for the coupling time of a Markov chain for non-trivial statistical-mechanics models. Using the "coupling from the past" protocol, this allows one to exactly sample the underlying equilibrium distribution. For spin glasses in two and three spatial dimensions, the local-patch algorithm works at lower temperatures than previous exact-sampling methods. We discuss variants of the algorithm which might allow one to reach, in three dimensions, the spin-glass transition temperature. The algorithm can be adapted to hard-sphere models. For two-dimensional hard disks, the algorithm allows us to draw exact samples at higher densities than previously possible.

Electronic version (arXiv)

Comment: Long version of the Chanal Krauth (2008) paper, containing an extension of the patch algorithm to hard spheres.

Python implementation of the patch algorithm

Personal tools