A method for estimating the number of hidden neurons in feed-forward neural networks based on information entropy

Computers and Electronics in Agriculture

Volumn 40, Issue 1-3, 2003, Pages 57-64

  Yuan, H C ^a,c   Xiong, F L ^b   Huai, X Y ^b  

^a UNIVERSITY OF SCIENCE AND TECHNOLOGY OF CHINA   (China)

^b INSTITUTE OF INTELLIGENT MACHINES   (China)

^c SHANGHAI OCEAN UNIVERSITY   (China)

Author keywords

Architecture selection; Hidden neurons; Information entropy; Neural network

Indexed keywords

AGRICULTURE; DECISION THEORY; NEURAL NETWORKS; TREES (MATHEMATICS);

HIDDEN NEURONS;

EXPERT SYSTEMS;

ARTIFICIAL NEURAL NETWORK;

EID: 0042740660     PISSN: 01681699     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0168-1699(03)00011-5     Document Type: Conference Paper

References (6)

1. Battiti, R., 1992. First- and second-order methods for learning: between steepest descent and Newton's method. Neural Comput. 4, 141-166.
2. Hertz, J., Krogh, A., Palmer, R.G., 1991. Introduction to the Theory of Neural Computation. Addison-Wesley, Redwood City, CA.
3. Li, Y.J., 1999. A method to directly estimate the number of the hidden neurons in the feedforward neural networks. J. Comput. 22 (11), 1204-1208.
4. Lin, G., 1991. Using multi-variable analysis method to evaluate the quality of Chinese tea and Japanese tea. Doctoral thesis of MingCheng University (in Japanese).
5. Setiono, R., 1995. A neural network construction algorithm which maximizes the likelihood function. Connect. Sci. 7 (2), 147-166.
6. Van Ooyen, A., Nienhuis, B., 1992. Improving the convergence of the backpropagation algorithm. Neural Networks 5 (3), 465-471.