Title :
A multilayer RBF network and its supervised learning
Author :
Chao, Jinhui ; Hoshino, Miho ; Kitamura, Tasuku ; Masuda, Takeshi
Author_Institution :
Dept. of Electr., Electron. & Commun. Eng., Chuo Univ., Tokyo, Japan
Abstract :
A general form of multilayer RBF networks is introduced. Complete supervised training rules for parameters are also presented. To achieve global convergence we apply a global optimization algorithm called the magic-brush method. This network can be naturally extended into a pyramid topology. Simulations show higher representation and generalization capability of the proposed networks comparing with the RBF and multilayer networks with sigmoid activation functions
Keywords :
convergence; learning (artificial intelligence); multilayer perceptrons; radial basis function networks; generalization capability; global convergence; global optimization algorithm; magic-brush method; multilayer RBF network; representation capability; supervised learning; supervised training rules; Chaotic communication; Clustering algorithms; Convergence; Costs; Hardware; Network topology; Nonhomogeneous media; Radial basis function networks; Shape; Supervised learning;
Conference_Titel :
Neural Networks, 2001. Proceedings. IJCNN '01. International Joint Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-7044-9
DOI :
10.1109/IJCNN.2001.938470
