A new algorithm of fuzzy neural networks with multiple form fuzzy weights
Lei Hongli, Jianbang Zhang, Zhang Dianzhi · 2003
A fuzzy neural network (FNN) architecture with fuzzy weight is proposed. The cost function depending on the fuzzy input and level sets of the fuzzy targets are defined, which can educe a learning algorithm that will be used to adjust the parameters which denote the fuzzy weight. The principles of the algorithm so that it can be applicable to the fuzzy weight with multiple forms are generalized. Then the algorithm is proved to be available by simulation studies.