So what exactly does the term neural network means ?

Well first thing first we usually say this term when we talk about deep learning or machine learning as a whole concept. The neural network is in fact could be taken as a way of understanding

these behemoths of words “MACHINE LEARNING / DEEP LEARNING”. In fact in more broader way these represents how we train computers to make decision on their own just like we do, using our intutions or past understanding or gut feeling.

logo

Just to make it easier for people you aren’t had background in biology, brain functions on sending electric pulse through these neurons, these neurons are actually the cells which keep or store the information in our brain. So what acutally happens that whenever we smell something new the neuron cells they fire up and store this particular smell and that way we remember the smell and its fragrance.

Now you understand how important is neurons in our decision making abilities they help us to identify, correct our path and help us in making decisions. Some what help us to minimise the error propagation. So the neural network is a network of these decision making small nodes which we call neurons. Each node represent a function that function can acutally help to correct the decision making abilities of our computer and machines.

We usually train machine on these network of nodes with datasets and over the period of time the machine’s abilites to attain correct answers for given input gets better and better as the neural network grows in complexity with obviously right data set.

Facisnating isn’t ?

These neural network are actually help the computers and machines to learn better and making full autonomus decision. We are in year 2020, we have seen it all automated cars, personality driven social network, and what not. These all are happening due to the rise of AI and these are all powered by the concepts such as neural network.

Now what I have represented is a very merry picture. As this sounds so easy it ain’t for sure but not hard to understand or work around these concepts. Infact with inflow of lib like tensorflow we can easliy deploy our net and run it for small to large complexities without sweat.