A LEARNING ALGORITHM OF FUZZY NEURAL NETWORK BASED ON TAKAGI-SUGENO'S MODEL
An Kai · Journal of Qufu Normal University · 2000
Taking normal functions as membership functions of the fuzzy neural network based on Takagi_Sugeno model, this paper gives a performance index used in learning and analysis its properties. Based on these results, an algorithm, called two-stage random search algorithm with Variable radius, is used in learning. A great many of examples show that the algorithm is simple and convenient and can make the fuzzy neural network obtain high precision. The definition of memory volume and the memory volume of a neural network based on Takagi_Sugeno model is given as well in this paper.