Title :
A study on three-layer perceptron with a capability of guaranteed learning
Author_Institution :
GoldStar Central Res. Lab., Seoul, South Korea
Abstract :
The author presents two types of three-layer perceptrons which are capable of guaranteed learning. In addition to perfect learning capability, the proposed structures contain only bipolar weights between layers, which turns out to be a significant improvement for the implementation process. The target value of an intermediate layer is determined by such a condition that binary input vectors are mapped into different positions in a linearly separable hyperspace.
Keywords :
learning (artificial intelligence); multilayer perceptrons; binary input vectors; bipolar weights; guaranteed learning; hyperspace; three-layer perceptron; Artificial neural networks; Backpropagation algorithms; Convergence; Energy resolution; Equations; Gold; Multilayer perceptrons; Neurons; Nonhomogeneous media; Vectors;
Conference_Titel :
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN :
0-7803-1421-2
DOI :
10.1109/IJCNN.1993.716803
