The different version of extremal optimization (EO) algorithm on two and three-dimensional spin glasses with Gaussian disorder was tested. It was shown that the jaded extremal optimization (JEO) extends the extremal optimization by adaptively reducing the frequency of flipping previously selected spins. It was found that the 2D ISG (Ising spin glass), using extremal optimization were consistent with those that would be seen for an NP-hard optimization problem with a similar local solution strategy. It was suggested that an algorithm such as JEO was useful to improve the configurations formed by whole sample crossover in genetic algorithms.
|Physical Review E - Statistical, Nonlinear, and Soft Matter Physics
|Published - May 2004
ASJC Scopus subject areas
- Statistical and Nonlinear Physics
- Statistics and Probability
- Condensed Matter Physics