|
An improved Particle Swarm Optimization (PSO) algorithm coupling with prior information for function approximation is proposed in present study. The prior information derived from the first-order derivative information of the approximated function is used to adjust the position of the particles. In the new approximation algorithm, feedforward neural network is first trained by improved PSO and then by BP. The prior information narrows the search space and guides the movement direction of the particles, so the convergence rate and the generalization performance are improved. Experimental results demonstrate that the new algorithm is more effective than traditional methods.
|
