Title :
Neural network initialization by combined classifiers
Author :
Van Breukelen, Martijn ; Duin, Robert P W
Author_Institution :
Dept. of Appl. Phys., Delft Univ. of Technol., Netherlands
Abstract :
If a set of linear classifiers in the same feature spaces is combined by a linear output classifier and if each of these classifiers has a sigmoid output-function then this set of classifiers has the same architecture as a feedforward neural network. A combined set of classifiers, however is trained in an entirely different way. In this paper it is shown that it can be advantageous to use such a set as an initialization for a neural network
Keywords :
neural nets; pattern classification; combined classifiers; feature spaces; feedforward neural network; linear output classifier; neural network initialization; sigmoid output-function; Backpropagation; Feedforward neural networks; Feedforward systems; Logistics; Neural networks; Neurons; Pattern recognition; Physics; Space technology;
Conference_Titel :
Pattern Recognition, 1998. Proceedings. Fourteenth International Conference on
Conference_Location :
Brisbane, Qld.
Print_ISBN :
0-8186-8512-3
DOI :
10.1109/ICPR.1998.711119
