Standing for adaptive linear element, as named by Bernard Widrow and Marcian Hoff, Adaline nets are structurally similar to Perceptron networks, but have a restricted range of -1 to 1. This range is necessary for learning based on minimizing the average squared error of the net. A derivative of this neural network, called the Madaline network, has more widespread use, but the Adaline is very adept at storing patterns, and can prove to be quite useful when a pattern-based decision is needed.

Log in or register to write something here or to contact authors.