Neural networks for pattern recognition and adaptive control
Date of Issue1999
School of Electrical and Electronic Engineering
This report presents a detail investigation on the pattern recognition ability of artificial neural networks and ways to use the inherent parallelism in the neural network to reduce the training time. In particular it presents a detailed study of parallel implementation aspects of Multilayer Feedforward Neural Networks used for handwritten character recognition. In this report the emphasis is on Training set parallelism.
DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Signal processing
Nanyang Technological University