VII. Generalized Simulated Annealing Algorithms Using Tsallis Statistics: Application to ±J Spin Glass Model

Abstract

Recently, the Tsallis generalized distribution of states has been adapted to various optimization algorithms. New Monte Carlo acceptance probabilities have been worked out and tested in a number of physical systems. In the paper we have applied the generalized simulated annealings to a ± spin glass of 70 X 70 spins on a square lattice. Since the system ground-state energy was known exactly the ability of the generalized acceptance probabilities to find low-energy states has been checked out. Efficiency of the new annealing procedures has also been compared with that of the traditional method with respect to the number of annealing steps.