论文信息 - Determining the number of hidden nodes by progressive training

Determining the number of hidden nodes by progressive training

An implementation of the back propagation (BP) scheme to train feedforward neural networks is presented. With the proposed implementation, the BP scheme can always itself determine the number of hidden nodes required to solve a particular problem. An illustration of the scheme is given.

Fu-Lai Chung | Tan Lee

[1] A. C. Tsoi. Multilayer perceptron trained using radial basis functions , 1989 .

[2] Kurt Hornik,et al. Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[3] Terrence J. Sejnowski,et al. Analysis of hidden units in a layered network trained to classify sonar targets , 1988, Neural Networks.

[4] Robert A. Jacobs,et al. Increased rates of convergence through learning rate adaptation , 1987, Neural Networks.

[5] Etienne Barnard,et al. A comparison between criterion functions for linear classifiers, with an application to neural nets , 1989, IEEE Trans. Syst. Man Cybern..

[6] Geoffrey E. Hinton,et al. Learning internal representations by error propagation , 1986 .

[7] Bernard Widrow,et al. Neural nets for adaptive filtering and adaptive pattern recognition , 1988, Computer.

[8] Frank Fallside,et al. An adaptive training algorithm for back propagation networks , 1987 .