An Online Self-constructing Fuzzy Neural Network with Restrictive Growth

In this chapter, a novel paradigm, termed online self-constructing fuzzy neural network with restrictive growth (OSFNNRG), which incorporates a pruning strategy into some restrictive growing criteria, is proposed. The proposed approach not only speeds up the online learning process but also results in a more parsimonious fuzzy neural network while maintaining comparable performance and accuracy by virtue of the growing and pruning mechanism. The OSFNNRG starts with no hidden neurons and parsimoniously generates new hidden units according to the new growing criteria as learning proceeds. In the second learning phase, all the free parameters of hidden units, regardless of whether they are new or originally existing, are updated by the extended Kalman filter (EKF) method. The performance of the OSFNNRG algorithm is compared with other popular algorithms in the areas of function approximation, nonlinear dynamic system identification and chaotic time series prediction, etc. Simulation results demonstrate that the proposed OSFNNRG algorithm is faster in learning speed and the network structure is more compact while maintaining comparable generalization performance and accuracy.