Chaotic maps were frequently introduced to generate random numbers and used to replace the pseudo-random numbers distributed in Gauss distribution in computer engineering. These improvements in optimization were called the chaotic improved optimization algorithm, most of them were reported better in literature. In this paper, we collected 19 classical maps which could all generate pseudo-random numbers in an interval between 0 and 1. Four types of chaotic improvement to original optimization algorithms were summarized and simulation experiments were carried out. The classical grey wolf optimization (GWO) and sine cosine (SC) algorithms were involved in these experiments. The final simulation results confirmed an uncertainty about the performance of improvements applied in different algorithms, different types of improvements, or benchmark functions. However, Results confirmed that Bernoulli map might be a better choice for most time. The code related to this paper is shared with https://gitee.com/lvqing323/chaotic-mapping.
Keywords: benchmark functions; chaotic improvements; chaotic maps; nature-inspired algorithms; simulation experiments.