Artificial neural network is a structure composed of artificial neurons or nodes which are connected one to another.
An artificial neuron receives one or more inputs that are separately weighted, and sums them all to produce an output.
The output is passed through a non-linear function, known as an activation function or transfer function.
The transfer functions usually have a sigmoid function shape, but they may be shaped as hyperbolic tangent function, and also take the form of other non-linear functions, piecewise linear functions, or step functions.
After preparing an artificial neuron network all properties are stabilized except for the weights.
Then adjustment to the weights occurs via model training, and when error is minimized to a predefined level the weights can be fixed and the model utilized.
These artificial networks are usually used for artificial intelligence (AI) and machine learning (ML) problems (predictive modeling, adaptive control).