A Neuron

The fundamental structural component of any neural network is a neuron. It is basically a model of the natural neuron present in the human nervous system. It can be mathematically described by a relation between an output and a set of inputs. The diagram depicting a neuron is shown below.

This means that theoretically, we can look at each neuron as composed of three logical units:
1. The synapses, or interconnections, each with a number wkj associated with it. This number multiplies the input coming to the neuron along this synapse. It is called the synaptic strength.
Note: The nomenclature is as follows: k refers to the k'thneuron, and j to the j'thinput from the previous layer. i.e. input along the synapse between the j'th input and the neuron is wkj xj
2. The linear combiner, or adder, which adds all the weighted inputs, along with a bias bk.

Sometimes bk can be looked upon as wk0, associated with another synapse at which a