Biological Motivation :

However remember there that in discussion first lecture is about how people have answered the question like: "How are we going to get an agent to proceed intelligently" then one of the answers was to realise which intelligence in individual humans is effected by our brains. In fact Neuro-scientists have told us that the brain is made up of architectures of networks of neurons and on the most basic level there are neurons can be seen as functions that were given some input and will either fire or not fire such depending on the nature of the input.

Moreover the input to certain neurons comes from the senses but there in generally the input to a neuron is a set of outputs from other neurons. Hence if the input to a neuron goes over a certain threshold after that the neuron will fire. Like same way one neuron firing will affect the firing of many other neurons so then information can be stored in terms of the thresholds set and the weight allocated by each neuron to each of its inputs. In "Artificial Neural Networks" (ANNs) are designed to mimic the behaviour of the brain. But there some ANNs are built into hardware it is the vast majority are simulated in software so we concentrate on these. However it's important not to take the analogy too farthis means that there really isn't much more similarity between artificial and animal neural networks. In fact particularly while the human brain is estimated to contain around 100,000,000,000 neurons, ANNs typically contain less than 1000 equivalent units.

However there the interconnection of neurons is much bigger in natural systems as well as the way in which ANNs store then manipulate information is a gross simplification of the way in where networks of neurons work in natural systems.