Designing Functional Networks Through Evolutionary Programming

Functional network is a recently introduced extension of neural networks. Like neural networks, nowadays, there is no system designing method for designing approximation functional networks structure. A new genetic programming designing modeling method, combining genetic programming and genetic algorithm, was proposed for hybrid identification of model structure and functional parameters by performing global optimal search in the complex solution space where the structures and parameters coexist and interact. These results also show that the proposed method in this paper can produce very compact network structure and the functional networks convergent precision are improved greatly