Title :
An improved learning algorithm for compact RBF networks
Author :
Lai, Xiaoping ; Li, Bin
Author_Institution :
Sch. of Inf. Eng., Shandong Univ., Weihai
Abstract :
An improved learning algorithm for RBF networks is presented in this paper. It allocates new neurons by a four-part novelty criterion, removes redundant neurons according to their error reduction rates, and updates output-layer weights by a recursive least-squares algorithm with Givens QR decomposition. Simulations on two benchmark problems demonstrate that the algorithm produces more compact networks than existing algorithms
Keywords :
learning (artificial intelligence); least mean squares methods; radial basis function networks; Givens QR decomposition; compact RBF networks; four-part novelty criterion; improved learning algorithm; recursive least-squares algorithm; Adaptive control; Electronic mail; Error correction; Function approximation; Least squares approximation; Neurons; Proposals; Radial basis function networks; Radio access networks; Resonance light scattering;
Conference_Titel :
Neural Networks and Brain, 2005. ICNN&B '05. International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-9422-4
DOI :
10.1109/ICNNB.2005.1614682
