artificial neural network (ANN)

Artificial neural networks are abstract mathematical models of brain structures and functions. An ANN (artificial neural network) is sometimes referred to as either a "connectionist system", "neurocomputer", or "PDP (parallel distributed processing)" model. An ANN system usually consists of a large collection of units where each unit has a scalar real-valued state which is called the unit's activation level. The activation levels of all units in the ANN system may also be arranged as elements of a vector. This vector is referred to as an activation pattern. A parameter of the ANN system which can be interpreted as describing the degree to which the activation level of one unit in the system influences the activation level of another unit is often referred to as a connection weight.

ANN’s have the ability to have knowledge of a thing never encountered before based on it’s similarities with things already known. ANN’s are also capable of complex function mapping and noise insensitivity. ANN’s are computer algorithms which consist of highly interconnected processing elements called neurons that produce either weak, strong, or intermediate signals based on the weighted sum of the input signals they receive. These neuron output signals are either the inputs for other nodes or the outputs of the ANN. One way the ANN obtains the correct outputs is by learning from a set of examples. Below illustrates a schematic of an ANN.

Most ANN systems are very complex high-dimensional nonlinear information processing systems. Unlike linear systems, closed form solutions to nonlinear information processing systems do not typically exist. On the other hand, a great deal of the qualitative characteristics of an ANN system's behavior can be analyzed and described using well-known engineering tools and techniques.