Artificial Neural Network Part 2


The Neuron: Neurons are the basic building block of Artificial Neural Network. Sole purpose of Artificial Neural Network is to mimic how human brain works. Human brain is an amazingly powerful biological machine which keeps evolving or keeps learning from past experiences. Human brain learns from the latest information available and changes decision accordingly. In ANN we also try to build algorithms with capability of learning on their own.

How Neurons work:



A neuron is a nerve cell that is the basic building block of the nervous system. Neurons are similar to other cells in the human body in a number of ways, but there is one key difference between neurons and other cells. Neurons are specialized to transmit information throughout the body.

These highly specialized nerve cells are responsible for communicating information in both chemical and electrical forms. In order to communicate, neurons need to transmit information both within the neuron and from one neuron to the next. This process utilizes both electrical signals as well as chemical messengers.

A single neuron is not of much use, like a single ant cannot do much but millions of ants together can build a whole colony or anthill. Similarly millions neurons working together while transmitting information from one neuron to the other do wonders. It is estimated that there are 100 billion (100,000,000,000) neurons (nerve cells) in the human brain. Signals can travel as slow as about 1 mph or as fast as about 268 mph. This amazing capability of neurons to transmit information at such a high speed makes it so special. Two main components which make transmission possible are dendrite and Axon.

Dendrite: Dendrites are the receiver of signal

Axon: Axons are transmitter of signal

Working of Artificial Neural Network:



As we can see, the green neuron is receiving input value from yellow neuron. It is similar to linear regression, where we have several independent variables put together into regression model and we get an output value. In the above given example, green neuron is receiving information from single input layer but the way neural networks work we usually have several hidden layers. The concept of information transformation remains same. Every hidden layer would transmit information to green neuron of the next layer as in our above example. This process of information transfer continues until we reach the outer layer.



Next Article: Artificial Neural Network Part 3



Please enter your comment!
Please enter your name here