A fast pruning algorithm for an Efficient Adaptive Fuzzy Neural Network
Juan Du, Meng Joo Er · 2010
A fast pruning algorithm for an Efficient Adaptive Fuzzy Neural Network (EAFNN) is presented in this paper. An EAFNN is a Takagi-Sugeno-Kang (TSK) type fuzzy model which is functionally equivalent to the Ellipsoidal Basis Function (EBF) neural network. An EAFNN uses the combined pruning algorithm where both Error Reduction Ratio (ERR) method and a modified Optimal Brain Surgeon (OBS) technology are used to remove the unneeded hidden units. Simulation works show the proposed pruning algorithm is very efficient. It can not only reduce the complexity of the network but also accelerate the learning speed.