Title :
Best match algorithm with radial basis functions (BMRBF)
Author :
Musavi, Mohamad T. ; Liu, W.J.
Author_Institution :
Dept. of Comput. Eng., Maine Univ., Orono, ME
Abstract :
Summary form only given, as follows. The authors discuss an algorithm for implementation of a best match radial basis function (BMRBF) neural network. The algorithm is based on the modified Kullback-Leibler information number with nearest-neighbor density estimate and the well-known supervised least mean square (LMS) method. The network architecture is a single hidden layer with RBF nodes. BMRBF is easy to implement and time-efficient and can be used for chaotic time series prediction and classification
Keywords :
filtering and prediction theory; neural nets; pattern recognition; time series; best match radial basis function neural net; chaotic time series prediction; classification; modified Kullback-Leibler information number; nearest-neighbor density estimate; pattern recognition; single hidden layer; supervised least mean square; Chaos; Computer networks; Least squares approximation; Nearest neighbor searches; Neural networks;
Conference_Titel :
Neural Networks, 1991., IJCNN-91-Seattle International Joint Conference on
Conference_Location :
Seattle, WA
Print_ISBN :
0-7803-0164-1
DOI :
10.1109/IJCNN.1991.155598
