1
00:00:02,300 --> 00:00:07,940
Here is our exclusive logical Agard explorer, and these are our inputs.

2
00:00:08,750 --> 00:00:12,740
This is a graph based on our inputs, which are X and Y.

3
00:00:14,880 --> 00:00:20,140
As we can see on this graph, we cannot separate them with only one line, they are not separated.

4
00:00:20,150 --> 00:00:22,710
Well, with line one line, there is no way to do that.

5
00:00:22,710 --> 00:00:26,220
So we are required to use two different lines.

6
00:00:26,490 --> 00:00:34,330
And we also learn in our artificial Narron that each line with each equation is representing one neuron.

7
00:00:34,560 --> 00:00:39,060
So how can we show it with the artificial neuron if we have two lines?

8
00:00:39,420 --> 00:00:45,740
In order to do that, let's just transfer them first on there or and and environment here.

9
00:00:45,750 --> 00:00:47,450
I'm just transferring them.

10
00:00:47,940 --> 00:00:50,300
We just took it and and all of them.

11
00:00:50,580 --> 00:00:53,250
And this is how they look like now.

12
00:00:53,250 --> 00:01:00,030
They are trackable, they are separate to bowl and we can already separate them with one line and this

13
00:01:00,030 --> 00:01:01,920
line we can write an equation for it.

14
00:01:01,920 --> 00:01:11,790
Let's say if this is Z2, we can show this is C one plus here is just passing from let's say if this

15
00:01:11,790 --> 00:01:22,950
is one, then it would be zero point five Z, one plus zero point five, meaning we can write an equation

16
00:01:22,950 --> 00:01:26,040
for that and we can represent it with one zero.

17
00:01:26,370 --> 00:01:29,660
So let's take a look at this block over there.

18
00:01:30,330 --> 00:01:31,560
These are of our inputs.

19
00:01:31,560 --> 00:01:37,650
The thing which we did here was first take the add in over of them in previous sessions, we already

20
00:01:37,650 --> 00:01:43,430
saw with order equations for and and or and how can we show them with a simple neuron.

21
00:01:43,800 --> 00:01:46,980
Now they are moving to another layer.

22
00:01:47,010 --> 00:01:50,450
This layer is our entry.

23
00:01:50,910 --> 00:02:00,120
Zebb one here we have two lines, each one and Z2 each one is X and Y Z to ease X or Y and finally we

24
00:02:00,120 --> 00:02:01,350
will have a Z three.

25
00:02:01,350 --> 00:02:12,990
We can show that this Z tree is equal to X, X or Y or simple.

26
00:02:12,990 --> 00:02:22,920
We we can already united as let's say, transfer all of them in one sine Z to minus one was zero point

27
00:02:22,920 --> 00:02:23,430
five.

28
00:02:24,210 --> 00:02:26,010
We D, X or gate.

29
00:02:26,190 --> 00:02:31,120
We saw the problem and we also provide a solution, a practical solution for it.

30
00:02:31,680 --> 00:02:38,400
This is a simple example of multilayer Perceptron, which we know he does MLP.

31
00:02:39,870 --> 00:02:46,390
We have this a structure and we will able to classify this a structure by making it into different layers.

32
00:02:46,650 --> 00:02:51,120
This is a basic fundamentals for multilayer perceptual.