Introduction to Machine Learning (ML)

What is machine learning?

 Before we start machine learning, let's look at some definitions. In this regard, Arthur Samuel said,

Field of study that gives computers the ability to learn without being explicitly programmed.

That is if the computer has a miraculous ability for which it can learn anything without writing a program on the subject beforehand.

Suppose a bipedal (humanoid or two-legged) robot can learn to walk on its own without any specific walking program, then it can be said that the robot has used a learning algorithm. We can easily write a program for a bipedal robot to walk. But that walk can't be called intelligent in any way, how intelligent is it if an embedded system is programmed to do just that particular thing? He can be called intelligent only if the behavior of the device changes with the change.

According to Tom Michel,

A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E.

Suddenly looking at the definition can be a bit of a problem, so it can be said through an example,

Suppose I built a machine that could play chess, then we could write the following parameters like this,

E = The machine played 500 complete sets of chess

T = Chess is the task of the machine

P = The machine does not win or lose the game

By definition,

If the number of games played by the machine increases (E) as well as its winning rate increases (P) then it must be understood that the machine is actually learning.

And it is absolutely impossible to do it through the program.