查看文章

Swarm-intelligence based algorithms attract researchers as they are simple and effective tools for solving highly non-linear and multi-modal real-world problems. Practical studies prove the curse of population size and its influence over swarm based algorithms. However, the specific measurements on the curse of population size and its impact are not so extensive. Literature reports varying the population size is also one kind of adaption mechanism to incorporate a well-balanced exploration and exploitation into any swarm-based algorithm. Therefore, the main focus of this study is measuring the impact of different constant population size over swarm based algorithms over different classes of functions and dimensions. Study has also attempted to find the correlation between population size, problem type, and dimension. Three well-known swarm-based algorithms, namely particle swarm optimization (PSO …