Original subtitles

1

Crais final step that we're going to look at noise goes into the generator generator generates dogs

2

with eyes this time and we can look more three dimensional I have ears for different.

3

Too bad they have too many eyes.

4

But some of them but already you know it's learned from the mistakes it made previously.

5

And so now we want to train the discriminator and we're going to need another batch of dog images we're

6

going to put them in.

7

And there you go see what outputs you've got.

8

We've got some valuable top to bottom and so what we can see now is that these values are slowly converging

9

so the value of getting closer the values for the dogs or the dogs are getting bigger with time and

10

it's harder it is becoming harder for the discriminator to discriminate between dogs and dogs.

11

I also noticed that in the Paduan retraining the general and that's good that means these images are

12

getting closer to what we want to.

13

More realistic dogs.

14

And again we're going to take these values Kellow can calculate the errors based on what we know they

15

should be they should be zeros at the top ones at the bottom.

16

And that error is going to be back propagated through the network of the discriminator.

17

Weights are going to be updated and then then it's time to train the generators.

18

Now we're going to train the generator.

19

We're going to have to put these images into the network.

20

And what happens next is we get an output.

21

You can see this output is lower than it was just now but also as you can see it slowly.

22

It's not as low as at the very start.

23

Again this is a sign it's showing us that these images are actually getting closer to realistic dog

24

images.

25

Again that these valves are going to be we're going to get the values collate the error and get back

26

Burkitt the error through the network of the generator and the weights there.

27

So there we go that's how it works.

28

These are just three steps in training in reality these like hundreds and thousands of these steps and

29

then multiple airports as well.

30

And as you can imagine through many many many many iterations these images are going to get better and

31

better and better and better through this struggle through this confrontation between the generator

32

and the discriminant.

33

Additional information is definitely available.

34

And one of the best place to go to is the original paper by in good fellow it's called generative adversarial

35

Annetts 2014 paper which you can find archive.

36

Yeah.

37

And he explains everything there.

38

And one more thing I wanted to mention here.

39

When we were talking about the generator we just said it's a neural network.

40

But this type of neural network we haven't discussed it is not even discussed in the annex.

41

It's called a deconvolution or neural network in that case you'll find artificial neural networks and

42

can we do illusional neural networks but not deconvolution all neural networks.

43

So I just wanted to make a quick note here so there is this illusional neural network what a deconvolution

44

all neural network is if you flip this upside down or run it back to front and then run it this way

45

instead of the normal way you run the other way.

46

Basically you start with the vector of values which is our Random signal.

47

Then at the end you'll get the image.

48

And if you'd like to learn more about decompositional neural networks there is some additional reading

49

over here.

50

We're not discussing it here because it's not the main concept of for us to focus on one focus on Ganns

51

But if you'd like to learn more about D-Conn. finance then this is a paper which talks about them is

52

called adaptive deconvolution networks for mid and high level feature learning by Mathieu's and others.

53

So yeah that's a go to people are forged continents hopefully you enjoyed this tutorial and you now

54

know how Ganns work in the background and then report seeing you back here next time.

55

And until then enjoy computer vision.