Choosing an Optimal Neural Network Size to Aid a Search through a Large Image Database

作者: K. Messer , J. Kittler

DOI: 10.5244/C.12.24

关键词:

摘要: In this paper a fast method of selecting neural network architecture for pattern recognition tasks is presented. We demonstrate that our proposed both input features and hidden neurons avoids the pitfalls exhibited by other methods reported in literature. It also shown resulting extremely lean while at same time significantly improving performance. The solution provides very useful tool which now being incorporated operations system used large image database surveys.

参考文章(13)
Matthew Kabrisky, Dennis W. Ruck, Steven K. Rogers, Wright-Patterson Afb, Feature Selection Using a Multilayer Perceptron ,(1990)
Christopher M. Bishop, Neural networks for pattern recognition ,(1995)
P. Pudil, J. Novovičová, J. Kittler, Floating search methods in feature selection Pattern Recognition Letters. ,vol. 15, pp. 1119- 1125 ,(1994) , 10.1016/0167-8655(94)90127-9
K Messer, J Kittler, M Kraaijveld, Selecting features for neural networks to aid an iconic search through an image database 6th International Conference on Image Processing and its Applications. ,vol. 1, pp. 428- 432 ,(1997) , 10.1049/CP:19970930
Scott E. Fahlman, Christian Lebiere, The Cascade-Correlation Learning Architecture neural information processing systems. ,vol. 2, pp. 524- 532 ,(1989)
Rudy Setiono, A penalty-function approach for pruning feedforward neural networks Neural Computation. ,vol. 9, pp. 185- 204 ,(1997) , 10.1162/NECO.1997.9.1.185
Yann LeCun, John Denker, Sara Solla, None, Optimal Brain Damage neural information processing systems. ,vol. 2, pp. 598- 605 ,(1989)
Rudy Setiono, Huan Liu, Neural-network feature selector IEEE Transactions on Neural Networks. ,vol. 8, pp. 654- 662 ,(1997) , 10.1109/72.572104
Babak Hassibi, David G. Stork, Second order derivatives for network pruning: Optimal Brain Surgeon neural information processing systems. ,vol. 5, pp. 164- 171 ,(1992)