Even on finite test sets smaller nets may perform better

Neural Networks

Volumn 10, Issue 2, 1997, Pages 369-385

  Elsken, Thomas ^a  

^a UNIVERSITY OF OSNABRÜCK   (Germany)

Author keywords

feedforward neural rats; identification from input output relation

Indexed keywords

ERRORS; LEARNING SYSTEMS; PERFORMANCE; SET THEORY; TRANSFER FUNCTIONS;

IDENTIFICATION FROM INPUT OUTPUT RELATION;

FEEDFORWARD NEURAL NETWORKS;

ARTICLE; ARTIFICIAL NEURAL NETWORK; PRIORITY JOURNAL;

EID: 0031104958     PISSN: 08936080     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0893-6080(96)00068-8     Document Type: Article

References (10)

1. Albertini, F., & Sontag, E. D. (1993). For neural networks, function determines form. Neural Networks, 6(7), 975-990.
2. Albertini, F., Sontag, E. D., & Maillot, V. (1993). Uniqueness of weights for neural networks, In R. Mammone (Ed.), Artitifical neural networks with applications in speech and vision. London: Chapman and Hall.
3. Baum, E. B., & Haussler, D. (1989). What size net gives valid generalization? In D. S. Touretzky (Ed.), Neural information processing systems I (pp. 81-90). Palo Alto, CA: Morgan Kaufmann.
4. Harp, S. A., Chamad, T., & Guha, A. (1990). Designing application specific neural networks using the genetic algorithm. In D. S. Touretzky (Ed.), Neural information processing systems II (pp. 447-454). Palo Alto, CA.: Morgan Kaufmann.
5. Hecht-Nielsen, R. (1989). Theory of the backpropagation neural network. Proceedings of the International Joint Conference on Neural Networks (pp. 593-605). New York: IEEE Publications.
6. Hertz, J., Krogh, H., & Palmer, R. G. (1991). Introduction to the theory of neural computation. Redwood City, CA: Addison-Wesley.
7. Kùrková, V., & Kainen, P. C. (1994). Functionally equivalent feedforward neural networks. Neural Computation, 6(3), 543-558.
8. Lang, S. (1977). Complex analysis. New York: Springer.
9. Schmitt, M. (1994). On the complexity of consistency problems. Ulmer Informatik Berichte, Nr. 94-01.
10. Sussmann, H. J. (1992). Uniqueness of the weights of minimal feedforward nets with a given input-output map. Neural Networks, 5(4), 589-593.