Finding the Most Influential Parameters of Coalitions in a PSO-CO Algorithm.

Literature reveals that optimization algorithms are generally composed of a large number of parameters that highly influence on its performance. In the early stages of the definition of a new algorithm, it is crucial to know how the uncertainty in the input parameters affects the behavior of the algorithm, influencing on its final output, so that it is possible to set up the most efficient configuration. In this work, we are making a sensitivity analysis using the Extended Fourier Amplitude Sensitivity Test to compute the first order effects and interactions for each parameter on a recently proposed particle swarm optimization algorithm that implements a dynamic structured swarm, based on coalitions. This technique, inherited from game theory, includes four new parameters that are analyzed and tested on a well-known benchmark for continuous optimization. Results give interesting insights of the importance of one of the parameters over the rest.