Introduction To Neural Network

Hello friends, in today’s blog post we are going to talk about Neural Networks of AI (Artificial Intelligence), in this post we will know in simple language what is Neural Network and how it works, so let’s start and know!

Neurons are connected to each other in very complex ways and all these connections are called Synapses. Now this must be coming in your mind that what is its relation with Artificial Neural Networks? So, now let us try to understand in a little deeper and easier way what neural networks are and how they learn.

What Are Neural Networks?

To understand Neural Network, we will take the example of our brain present in our body, so let’s start, so whenever we read or write something, do you know which part of your body is working at that time, like you are reading this post at this time, do you know which part of your body is doing the work of thinking?

Although I think you must be knowing, but still let me tell you that this work is done by our brain. But have you ever wondered how our brain works? So let us tell you that there are many small cells present inside the brain, which we know as Neurons.

These are the same neurons that collect information from the environment around us, and later understand it well, and then send the same information to other neurons so that our body continues to function properly without any problem.

Artificial Neural Networks

Artificial Neural Networks (ANNs) contain many small artificial neurons, which we know as units. All these units are arranged in many different layers and then later all of them come together to form a complete neural network.

The number of units in its layers is not always equal, that is, the number of units present in it can always be different. Sometimes it happens that only a few units are present at some place, while sometimes lakhs of units can be present simultaneously on a ship. The number of units depends on how complex the network you are building has to learn.

Layers Of Artificial Neural Network

Well, if seen, usually only three types of layers are present in any Artificial Neural Network:

Input Layer: This is the layer whose job is to receive any information coming from outside.
Hidden Layers: This is the same layer in which the data received by the input layer is processed. Apart from this, one or more hidden layers may be present in any network.
Output Layer: The job of this layer is to give a final output according to all the processes done before it. That is, in simple language, this layer prepares the response of a network.

Let us now understand this entire process in a simple language, whenever the input layer receives any kind of data, it passes through hidden layers. Then after that the hidden layers transform that input, so that its output layer can understand it correctly and well and then finally provide a correct output.

Units And Connections Of Artificial Neural Networks (ANNs)

In these Artificial Neural Networks (ANNs), generally the units of each layer are connected to the units of some other layer. Whenever connections are made by combining the units of these layers, then all those connections have their own weight, whose job is to decide how much effect any one unit will have on another unit.

Whenever data is transferred from one unit to another, the Neural Network starts understanding the pattern of that data. At the time of training, the weights of all these layers are adjusted in advance so that the network can learn even better so that the output obtained from it is much more accurate.

Conclusion

So friends, in today’s blog post we have understood in easy and simple language what Neural Networks are and how they work. With the help of this blog, we learned that just as there are connections between neurons in our brain, similarly in Artificial Neural Networks (ANNs) too many units and layers are interconnected.

I hope you liked this information and learned something new from it. If yes, then definitely share this post with your friends also. See you in the next post, till then thanks!

Read Also: