This content was published by Andrew Tomazos and written by several hundred members of the former Internet Knowledge Base project.

Neural Networks

(Artificial) Neural Networks (ANN) are programs which attempt to emulate the way a vertebrate brain stores and recalls information. The software consists of many (thousands or even millions of) data cells (nodes or neurons) which, when stimulated (by input), signal the nodes to which they are connected. In some ANNs all nodes are connected to every other node. Each connection has a value, or weight, which determines the strength of the signal reaching the receiving node from the sending node. Stimulating several nodes at once (with a pattern of input) causes the learning algorithms to modify the weight values which results in the development of associative memory. One advantage is that after learning has taken place, incomplete or inexact input can cause recall of the complete original pattern.

Back to Index