FIGURE 7.3-2 The ADALINE's decision boundaries used to separate the «s from the os.
Three-layer ANNs with multiple output neurons can be made from ADALINES. The w3 weights of the output layer can be optimized or trained by use of the Widrow-Hoff LMS algorithm, described below. A back-propagation algorithm is generally required to train the w2 and Wj weights of the hidden layers (Widrow and Winter, 1988).
The LMS training algorithm for a single MADALINE adjusts w to minimize the mean squared error (MSE) of the classifier. The MSE can be written in matrix notation (Hecht-Nielsen, 1990):
Was this article helpful?
This guide will help millions of people understand this condition so that they can take control of their lives and make informed decisions. The ebook covers information on a vast number of different types of neuropathy. In addition, it will be a useful resource for their families, caregivers, and health care providers.