Enhancing Monte Carlo methods by using a generalized fluctuation theory
Statistical Mechanics
2007-05-23 v1
Abstract
According to the recently obtained thermodynamic uncertainty relation, the microcanonical regions with a negative heat capacity can be accessed within a canonical-like description by using a thermostat with a fluctuating inverse temperature. This far-reaching conclusion is used in this Letter for enhancing the potentialities of the well-known Swendsen-Wang cluster algorithm in order to access to the anomalous microcanonical states of the q=10 state Potts model on a square lattice, which exhibits a first-order phase transition in its thermodynamical description.
Cite
@article{arxiv.cond-mat/0612683,
title = {Enhancing Monte Carlo methods by using a generalized fluctuation theory},
author = {L. Velazquez},
journal= {arXiv preprint arXiv:cond-mat/0612683},
year = {2007}
}
Comments
RevTex4 style. 4 pages, 1 eps figure