Geometrical Selection of Important Inputs with Feedforward Neural Networks
In this paper, we introduce a method that allows to evaluate efficiently the ‘importance’ of each coordinate of the input vector of a neural network. This measurement can be used to obtain information about the studied data. It can also be used to suppress irrelevant inputs in order to speed up the classification process conducted by the network.
KeywordsMutual Information Input Vector Feedforward Neural Network Wavelet Network Irrelevant Input
Unable to display preview. Download preview PDF.
- C. Gégout, B. Girau and F. Rossi. Generic Back-Propagation in Arbitrary Feedforward Neural Networks. In D.W. Pearson, N.C. Steele, and R.F. Albrecht, editors, Int. Conf. on Artificial Neural Networks and Genetic Algorithms, pages 168–171, Als, April 1995. Springer Verlag.Google Scholar
- A.J. Miller. Subset Selection in Regression. Chapman and Hall, 1990.Google Scholar
- W.H. Press, S.A. Teukolsky, W.T. Vetterling, and B.P. Flannery. Numerical Recipes in C. Cambridge University Press, second edition, 1992.Google Scholar
- F. Rossi. Attribute suppression with multi-layer perceptron. In CESA Multiconference, Symposium on Robotics and Cybernetics, pages 542–547, Lille-France, July 1996. IMACS.Google Scholar