Geometrical synthesis of MLP neural networks
Neurocomputing
NN'07 Proceedings of the 8th Conference on 8th WSEAS International Conference on Neural Networks - Volume 8
NN'07 Proceedings of the 8th Conference on 8th WSEAS International Conference on Neural Networks - Volume 8
A neural network structure with constant weights to implement convex recursive deletion regions
NN'07 Proceedings of the 8th Conference on 8th WSEAS International Conference on Neural Networks - Volume 8
On implementation of nested rectangular decision regions by multi-layer perceptrons I: algorithm
NN'06 Proceedings of the 7th WSEAS International Conference on Neural Networks
NN'06 Proceedings of the 7th WSEAS International Conference on Neural Networks
NN'06 Proceedings of the 7th WSEAS International Conference on Neural Networks
Implementation feasibility of convex recursive deletion regions using multi-layer perceptrons
WSEAS Transactions on Computers
An empirical improvement of the accuracy of RBF networks
Proceedings of the 2nd International Conference on Interaction Sciences: Information Technology, Culture and Human
Empirical determination of sample sizes for multi-layer perceptrons by simple RBF networks
WSEAS Transactions on Computers
Hi-index | 0.00 |
A sufficient condition that a region be classifiable by a two-layer feedforward neural net (a two-layer perceptron) using threshold activation functions is that either it be a convex polytope or that intersected with the complement of a convex polytope in its interior, or that intersected with the complement of a convex polytope in its interior or... recursively. These have been called convex recursive deletion (CoRD) regions. We give a simple algorithm for finding the weights and thresholds in both layers for a feedforward net that implements such a region. The results of this work help in understanding the relationship between the decision region of a perceptron and its corresponding geometry in input space. Our construction extends in a simple way to the case that the decision region is the disjoint union of CoRD regions (requiring three layers). Therefore this work also helps in understanding how many neurons are needed in the second layer of a general three-layer network. In the event that the decision region of a network is known and is the union of CoRD regions, our results enable the calculation of the weights and thresholds of the implementing network directly and rapidly without the need for thousands of backpropagation iterations