Title :
A fast pruning algorithm for an Efficient Adaptive Fuzzy Neural Network
Author :
Du Juan ; Er Meng Joo
Author_Institution :
Sch. of Electr. & Electron. Eng., Nanyang Technol. Univ., Singapore, Singapore
Abstract :
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.
Keywords :
adaptive systems; decision trees; fuzzy neural nets; learning (artificial intelligence); Optimal Brain Surgeon technology; Takagi-Sugeno-Kang type fuzzy model; efficient adaptive fuzzy neural network; ellipsoidal basis function neural network; error reduction ratio method; fast pruning algorithm; Adaptive control; Adaptive systems; Automatic control; Equations; Erbium; Fuzzy control; Fuzzy neural networks; Fuzzy systems; Programmable control; Training data;
Conference_Titel :
Control and Automation (ICCA), 2010 8th IEEE International Conference on
Conference_Location :
Xiamen
Print_ISBN :
978-1-4244-5195-1
Electronic_ISBN :
1948-3449
DOI :
10.1109/ICCA.2010.5524417