Title :
On the properties of the feedforward method: A simple training law for on-chip learning
Author :
Petridis, Vassilios ; Paraschidis, Kyriazis
Author_Institution :
Dept. of Electr. Eng., Aristotelian Univ. of Thessaloniki, Greece
fDate :
11/1/1995 12:00:00 AM
Abstract :
This paper investigates the properties of the so-called feedforward method, which is a very simple training law suitable for on-chip learning. Its merit is conceptual and implementational simplicity. Its signals do not propagate in both directions and it works for various types of activation function, a feature that makes it particularly effective in the case of unmodeled activation functions. Extensive simulation has shown that this method is usually faster than backpropagation
Keywords :
feedforward neural nets; iterative methods; learning (artificial intelligence); search problems; activation function; analogue digital network; feedforward neural networks; iterative method; multidimensional search; on-chip learning; Artificial neural networks; Backpropagation; Computational modeling; Computer architecture; Digital circuits; Gradient methods; Kalman filters; Neural network hardware; Newton method; Perturbation methods;
Journal_Title :
Neural Networks, IEEE Transactions on
