DocumentCode :
2539312
Title :
A novel Online Self-organizing Fuzzy Neural Network for function approximation
Author :
Wang, Ning ; Qiu, Chidong ; Niu, Xiaobing ; Xue, Zhengyu
Author_Institution :
Marine Eng. Coll., Dalian Maritime Univ., Dalian, China
fYear :
2010
fDate :
7-9 July 2010
Firstpage :
550
Lastpage :
555
Abstract :
In this paper, we propose a novel Online Self-constructing Fuzzy Neural Network (OSFNN) which extends the ellipsoidal basis function (EBF)-based fuzzy neural networks (FNNs) by permitting input variables to be modeled by dissymmetrical Gaussian functions (DGFs). Due to the flexibility and dissymmetry of left and right widths of the DGF, the partitioning made by DGFs in the input space is more flexible and more economical, and therefore results in a parsimonious FNN with high performance under the online learning algorithm. The geometric growing criteria and the error reduction ratio (ERR) method are used as growing and pruning strategies respectively to realize the structure learning algorithm which implements an optimal and compact network structure. The proposed OSFNN starts with no hidden neurons and does not need to partition the input space a priori. In addition, all the free parameters in premises and consequents are adjusted online based on the ε-completeness of fuzzy rules and the linear least square (LLS) approach, respectively. The performance of the proposed OSFNN paradigm is compared with other well-known algorithms like ANFIS, OLS, GDFNN, SOFNN and FAOS-PFNN, etc., on a benchmark problem in the field of function approximation. Simulation results demonstrate that the proposed OSFNN approach can facilitate a more powerful and more economical FNN with better performance of approximation and generalization.
Keywords :
Gaussian processes; function approximation; fuzzy neural nets; fuzzy set theory; generalisation (artificial intelligence); learning (artificial intelligence); least squares approximations; parameter estimation; radial basis function networks; dissymmetrical Gaussian function; ellipsoidal basis function; error reduction ratio; function approximation; fuzzy neural network; linear least square method; online learning algorithm; online self organizing network; pruning strategy; structure learning algorithm; Approximation algorithms; Function approximation; Fuzzy neural networks; Input variables; Least squares approximation; Neurons; Training;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Cognitive Informatics (ICCI), 2010 9th IEEE International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-8041-8
Type :
conf
DOI :
10.1109/COGINF.2010.5599680
Filename :
5599680
Link To Document :
بازگشت