The ANN is a collection of artificial neurons that perform simple operations on the data; the output from this is passed to another neuron. The output generated at each neuron is called its activation function. An example of a multilayer perceptron model can be seen in the following screenshot:
Each link in the preceding figure is associated to weights processed by a neuron. Each neuron can be looked at as a processing unit that takes input processing and the output is passed to the next layer, as shown in the following screenshot:
The preceding figure demonstrates three inputs combined at neuron to give an output that may be further passed to another neuron. The processing conducted at the neuron could be a very simple operation such as the input multiplied by weights followed by summation or a transformation operation such as the sigmoid activation function.