Improvements on Niche Genetic Algorithm

In order to avoid premature convergence and occurance of minimal deceptive problems, an improved niche genetic algorithm (NGA) is presented. The algorithm is based on the adaptive mutation operator and crossover operator that adjusts the crossover rate and frequency of mutation of each individual, and adopts the gradient of the individual to decide their mutation value. This approach is used in Shubert function optimization. Through comparisons to GA and NGA, the result of improved algorithms shows its feasibility and effectivity.