Optimal size of a feedforward neural network: How much does it matter?

Proceedings of the Joint International Conference on Autonomic and Autonomous Systems and International Conference on Networking and Services, ICAS/ICNS 2005

Volumn 2005, Issue , 2005, Pages 69-

  Lipo, Wang ^a   Hou, Chai Quek ^b   Keng, Hoe Tee ^b   Nina, Zhou ^b   Chunru, Wan ^b  

^a XIANGTAN UNIVERSITY   (China)

^b NANYANG TECHNOLOGICAL UNIVERSITY   (Singapore)

Author keywords

Hidden neurons; Learning; Neural networks; Occam's Razor

Indexed keywords

COMPUTER NETWORKS; COMPUTER SIMULATION; ERROR ANALYSIS;

OCCAM'S RAZOR; VALIDATION ERRORS;

FEEDFORWARD NEURAL NETWORKS;

EID: 33845330227     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/ICAS-ICNS.2005.72     Document Type: Conference Paper

References (10)

1. Y. Abu-Mostafa. The vapnik-chervonenkis dimension: Information versus complexity in learning. Neural Computation, 1(3):312-317, 1998.
2. H. Akaike. Statistical predictor identification. Annals of the Institute for Statistical Mathematics, 22:203-217, 1970.
3. H. Akaike. Information theory and an extention of the maximum likelihood principle. B.N. Petrov and F. Csaki, eds. Proceedings 2nd International Symposium on Information Theory, pages 267-281, 1973.
4. P. Bartlett. Vapnik-chervonenkis dimension bounds for two-and three-layer networks. Neural Computation, 5(3):371-373, 1993.
5. R. Caruana. Generalization vs. net size, Neural Information Processing Systems. Tutorial, 1993.
6. S. Lawrence, C. Giles, and A. Tsoi. What size neural network gives optimal generalization? convergence properties of backpropagation. University of Maryland Technical Report, UMIACS-TR-96-22, 1996.
7. W. Maass. Vapnik-chervonenkis dimension of neural networks, in M.A. Arbib, ed., The Handbook of Brain Theory and Neural Networks, 1995.
8. J. Moody. The effective number of parameters: An analysis of generalization and regularization learning systems, in J. Moody, S.J. Hanson, and R.P. Lippmann, eds., Advances in Neural Information Processing, 4:847-854, 1992.
9. K. Muller, M. Finke, K. Schulten, N. Murata, and S. Amari. A numerical study on learning stochastic multi-layer feedforward networks. Neural Computation, 1996.
10. B. Ripley. Statistical ideas for selecting network architectures. Invited Presentation, Neural Information Processing Systems 8, 1995.