Computer Science and Information Engineering, World Congress on (2009)
Los Angeles, California USA
Mar. 31, 2009 to Apr. 2, 2009
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/CSIE.2009.76
Being difficult to determine hidden units’s number and unsuitable to select central position in radial basis function (RBF) layer, Particle Swarm Optimization and Resource Allocation (RAN) were proposed for training RBF neural networks. First, determine units’s number in RBF layer using RAN. Then, optimize RBF parameters such as central position, width and weights based on PSO. The simulation results show that the new method has better approximation ability, the shorter time and the higher precision.
Man Chun-tao, Zhang Li-yong, Wang Kun, "A New Training Algorithm for RBF Neural Network Based on PSO and Simulation Study", Computer Science and Information Engineering, World Congress on, vol. 04, no. , pp. 641-645, 2009, doi:10.1109/CSIE.2009.76