Before learning TensorFlow, we should know what it is. TensorFlow is a machine learning or deep learning framework.
Who has created TensorFlow?
Now, as a beginner you may wonder, what is machine learning and what is deep learning?
Are they the same?
Or do they relate to Artificial Intelligence or AI?
Well, these questions are natural.
Because Artificial Intelligence, Machine Learning and Deep Learning or Neural Networks relate to one another.
But what do these terms actually mean and how do they relate to each other?
Let’s try to understand Artificial Intelligence first.
Firstly, Artificial Intelligence or AI is a field of computer science.
Secondly, AI aims to give computers human intelligence.
Certainly, AI has many approaches to reach its goal.
And that includes machine learning and deep learning.
Therefore, we can think, machine learning and deep learning as sub fields of AI.
However, if we say that, we will oversimplify the whole topic.
Still, for a beginner, this definition seems to be meaningful.
What is TensorFlow and Machine Learning?
Machine Learning is a set of related techniques through which we train computers.
Why?
Because we want the computers to perform a particular task rather than by explicitly programming them.
Now, it means many things.
In classical programming we write code or instructions so that computers can execute them. Right?
But in machine learning just the opposite takes place.
Here we give computers machine learning algorithms so that computers can give us instructions what to do, or what not to do.
But in this scenario, what is the position of deep learning or the Neural Network?
Let’s try to understand that.
Basically our brains have billions of neurons or nerve cells that have a very complex network to communicate with each other.
We cannot do the exact same thing with the help of machine learning.
But the main inspiration comes from that angle, of course.
Therefore we name it Deep Learning.
Why?
Because deep learning is a subfield of machine learning that uses multi-layered neural networks.
We just said, the main inspiration has come from the human neural network.
As a result, often we use “machine learning” and “deep learning” interchangeably.
In addition, machine learning and deep learning also have many subfields, branches, and special techniques.
As we progress, we will see how Supervised Learning and Unsupervised Learning are different from each other.
In short, in supervised learning we know what we want to teach the computer.
But in unsupervised learning the computer figures out what it can learn.
Quite naturally, supervised learning is the most common type of machine learning.
As a beginner we will concentrate on that.
Later we will also learn unsupervised learning.
So stay tuned.
Leave a Reply