One of the first ANNs, the Perceptron, was developed by Rosenblatt (1958; 1962). His Mark I perceptron was a pattern recognition system that could learn to recognize simple visual patterns presented to its "compound eye" receptor plane, which contained 400 CdS photosensors in a square, 20 x 20 array. This perceptron had available a total of 512 connections with adjustable weights. The weights were varied by motor-driven potentiometers. The connections could be set by a patch panel, and were usually connected "randomly." The training law sent signals to the servomotors that adjusted the potentiometers. An ingenious but clumsy system by today's standards.
In its simplest form, the Rosenblatt perception was a one-layer system (for some reason, the input nodes are not counted as a layer); 400 receptors in the input layer, and one output "neuron" is in the output layer. This perceptron was a simple binary classifier; the output of the neuron was taken as zero if the sum of weighted inputs plus the bias input was < 0; it was "1" if the bias plus weighted inputs was S 0. Mathematically:
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.