114 Comparison between new optimization method using a probability theory and other optimization methods

Abstract

In this paper, we compared efficiency of Stochastic optimization (SO) with other optimization methods, i.e., Genetic algorithm (GA) and Differential evolution (DE). For the index of efficiency, we compared function call numbers until we could obtain an approximate solution of some famous test functions. The results showed that SO could obtain approximate solutions inefficiently than DE and effectively than GA. The results also showed that SO is an optimization method that is easy to deal with, because SO needed not adjust parameters for each problem unlike GA and DE.