Simulated annealing and Boltzmann machines: a stochastic approach to combinatorial optimization and neural computing
Genetic algorithms + data structures = evolution programs (3rd ed.)
Genetic algorithms + data structures = evolution programs (3rd ed.)
Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
| Hi-index | 0.00 |
Genetic algorithms have good global search ability and relatively fast convergence rate. Micro- GA does not need to adopt mutation, for it introduces completely new individuals in the mating pool that have no relation to evolved close individuals. The pool size is smaller than those of the simple GA cases, which need a big pool to achieve much variety in individuals. Sometimes, these genetic algorithms are hard to pinpoint the optimal solution that is correct up to several significant digits. They can approach fast to the vicinity of the global optimum, but, thereafter, march too slowly to it in many cases. To enhance the convergence rate, hybrid methods have been developed. The SZGA (Successive Zooming Genetic Algorithm) zooms the search domain for a specified number of times to obtain the optimal solution correct up to several significant digits. It has three parameters controlling the computational procedure that affect on the efficiency of entire optimization. We optimized the SZGA itself to obtain the optimal values of the parameters that satisfy reliability and accuracy condition and minimize the computational quantity. The optimal values of the three parameters in the successive zooming genetic algorithm were obtained for different numbers of design variables that assure target reliability of 99.9999% and target accuracy of 10-6. The SZGA with optimal parameters was applied to many test functions and actual problems, successively and effectively.