Title :
Geometrical learning algorithm for multilayer neural networks in a binary field
Author :
Park, Sung-Kwon ; Kim, Jung H.
Author_Institution :
Dept. of Electron. Commun. Eng., Hanyang Univ., Seoul, South Korea
fDate :
8/1/1993 12:00:00 AM
Abstract :
A geometrical expansion learning algorithm for multilayer neural networks using unipolar binary neurons with integer connection weights, which guarantees convergence for any Boolean function, is introduced. Neurons in the hidden layer develop as necessary without supervision. In addition, the computational amount is much less than that of the backpropagation algorithm
Keywords :
Boolean functions; feedforward neural nets; learning (artificial intelligence); Boolean function; binary field; geometrical learning algorithm; hidden layer; integer connection weights; multilayer neural networks; unipolar binary neurons; Backpropagation algorithms; Fault diagnosis; Hypercubes; IEEE Computer Society Press; Intelligent networks; Interconnected systems; Multi-layer neural network; Multiprocessing systems; Neural networks; Neurons;
Journal_Title :
Computers, IEEE Transactions on
