No abstract

The back-propagation(BP) dgorithm is widely used for finding optimum weights of multi-layer neural networks in many pattern recognition applications. However, the critical drawback of the BP algorithm is its slow convergence of error. The major reason for this slow convergence is the "premature saturation" which is a phenomenon that the error of a neural network stays almost constant for some period of time during learning. It is known to be caused by an inappropriate set of initial weights. In this paper, the probability of premature saturation in BP algorithm has been derived in terms of the maximum value of initial weights, the number of nodes in each layer. and the maximum slope of sigmoidal activation function; it has been verified by Monte Carlo simulation. Using this result, the premature saturation can be avoided with proper initial weight settings.

No abstract

No abstract