No abstract

No abstract

No abstract

AESTRACT Two conditions for reducing the number of learning iterations in back-propagation artificial neural network are introduced in this paper. The first condition is to scale the target output so that it falls within a small range 20.1 of the point at which the slope of tbe nonlinear activation function of the output node is maximum. This point is 0.5 for the sigmoid function.The second condition is to learn the input patterns selectively not sequentially till the error is reduced below the desired limit. Introducing these two new techniques does not effect the memory retention or generalization capabilities of such networks. Application of these concepts to the classical XOR problem, resulted in a reduction in the number of learning iterations by a factor of 7 over the results published by Rumelhart.

No abstract