WEBVTT

00:00.000 --> 00:07.000
In this video, we will recall the basics of ML.

00:07.000 --> 00:13.000
Understand the types of ML.

00:13.000 --> 00:19.000
Recalling the basics of ML, machine learning is when computers learn from data and improve

00:19.000 --> 00:22.000
on their own without being directly programmed.

00:22.000 --> 00:29.000
There are three main types, supervised learning, unsupervised learning, reinforcement learning.

00:29.000 --> 00:34.000
The computer learns from labeled data.

00:34.000 --> 00:39.000
Where the correct answers are already given, the computer studies the shape and color.

00:39.000 --> 00:44.000
In the same way, this type of learning is also used in emails.

00:44.000 --> 00:48.000
Emails can be labeled as spam or not spam.

00:48.000 --> 00:54.000
The computer learns from these labels and then predicts which new emails are spam.

00:54.000 --> 00:57.000
Unsupervised learning.

00:57.000 --> 01:02.000
The computer gets data but no labels, so it has to group things on its own.

01:02.000 --> 01:03.000
Example.

01:03.000 --> 01:07.000
You give it a bunch of fruit pictures without telling the names.

01:07.000 --> 01:09.000
The computer notices similarities.

01:09.000 --> 01:14.000
Round red ones go in one group, long yellow ones in another.

01:14.000 --> 01:19.000
It doesn't know the names, but it finds patterns.

01:19.000 --> 01:21.000
Reinforcement learning.

01:21.000 --> 01:26.000
The computer learns by trial and error, getting rewards for good moves and penalties for bad ones.

01:26.000 --> 01:30.000
Similarly, this type of learning is used in self-driving cars.

01:30.000 --> 01:37.000
Where the car is rewarded for safe moves and penalized for mistakes until it learns how to drive correctly.

01:37.000 --> 01:45.000
In short, supervised learning works with answers, unsupervised finds hidden patterns, and reinforcement learns from rewards.

01:45.000 --> 01:52.000
Together, these three types make computers smarter and help in many real-life applications.