作者: Bernd A Berg , Alain Billoire , Wolfhard Janke
DOI: 10.1016/S0378-4371(02)01751-X
关键词: Statistical physics 、 Mathematics 、 Sampling (statistics) 、 Boltzmann constant 、 Orders of magnitude (time) 、 Ising model 、 Spin glass 、 Replica 、 Monte Carlo method 、 Order statistic
摘要: Abstract We present results of recent high-statistics Monte Carlo simulations the Edwards–Anderson Ising spin-glass model in three dimensions. The study is based on a non-Boltzmann sampling technique, multi-self-overlap algorithm which specifically tailored for rare-event states. thus concentrate those properties are difficult to obtain with standard canonical Boltzmann such as free-energy barriers F B q probability densities P J ( ) Parisi overlap parameter and behavior tails disorder averaged density )=[ )] av . Our disagree mean-field predictions support extreme order statistics over many orders magnitude.