What is Machine Learning?

Two definitions of Machine Learning are offered. Arthur Samuel described it as: "the field of study that gives computers the ability to learn withoute being explicitly programmed. "This is an older, informal definition.

​

Tom Mitchell provides a more modern definition: "A computer program is said to learn from experience E with repect 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."

​

Example: playing checkers.

E = The experience of playing many games of checkers
T = The task of playing checkers
P = The probability that the program will win the next game

---

Let's see a question:

​

"A computer program is said to learn from experience E with repect to task T and some performance measure P, If its performance on T, as measured by P, improves with experience E."

Suppose your email program watches which emails you do or do not mark as spam, and based on that learns bow to better filter spam. classify belows as E or T or P.

​

1. Classify emails as spam or not spam.

2. Watching you label emails as spam or not spam.

3. The number(or fraction) of emails correctly classfied as spam/not spam.

​

The first is T, because this is real activity.

The second is E, because you can learn how to classify emails as spam or not spam from this work.

The third is P, because this is a standard of performance.

---

In general, any machine learning problem can be assigned to one of two broad classfications:

Supervised learning  VS  Unsupervised learning

And there are some other algorithms based on machine learning: Reinforcement learning, recommender system etc. But the main algorithms are Supervised and Unsupervised learning.