1
00:00:00,000 --> 00:00:03,449
Hello everyone! And welcome to the deep

2
00:00:03,449 --> 00:00:05,430
learning fundamentals with Keras course.

3
00:00:05,430 --> 00:00:07,710
I am Alex Aklson, your course

4
00:00:07,710 --> 00:00:10,559
instructor. In this introductory course, I

5
00:00:10,559 --> 00:00:13,349
will try to teach you the basics of one

6
00:00:13,349 --> 00:00:15,269
of the hottest topics in the field of

7
00:00:15,269 --> 00:00:18,930
data science, which is deep learning. This

8
00:00:18,930 --> 00:00:20,490
course will consist of four modules,

9
00:00:20,490 --> 00:00:22,890
which you should be able to complete in

10
00:00:22,890 --> 00:00:26,849
four weeks. In module 1, I will try to

11
00:00:26,849 --> 00:00:29,310
motivate you about this field. I know

12
00:00:29,310 --> 00:00:31,500
that the fact that you are here, you must

13
00:00:31,500 --> 00:00:33,630
be already motivated to learn about deep

14
00:00:33,630 --> 00:00:35,910
learning, but I will share with you some

15
00:00:35,910 --> 00:00:38,250
very exciting applications of deep

16
00:00:38,250 --> 00:00:40,350
learning that will hopefully motivate

17
00:00:40,350 --> 00:00:42,989
you even more, and demonstrate to you

18
00:00:42,989 --> 00:00:44,940
that there is almost no limit to what

19
00:00:44,940 --> 00:00:47,550
can be done using deep learning. We will

20
00:00:47,550 --> 00:00:49,829
also briefly cover neurons and neural

21
00:00:49,829 --> 00:00:51,780
networks in the brain, so you can

22
00:00:51,780 --> 00:00:54,449
appreciate how they inspire artificial

23
00:00:54,449 --> 00:00:56,969
neural networks. Then, we will start

24
00:00:56,969 --> 00:00:59,010
learning about the different topics

25
00:00:59,010 --> 00:01:00,719
associated with artificial neural

26
00:01:00,719 --> 00:01:03,210
networks, and in this module we will

27
00:01:03,210 --> 00:01:05,069
focus on the process of forward

28
00:01:05,069 --> 00:01:10,530
propagation. In module 2, we will continue

29
00:01:10,530 --> 00:01:12,060
learning about artificial neural

30
00:01:12,060 --> 00:01:15,030
networks, and focus on how an artificial

31
00:01:15,030 --> 00:01:18,659
neural network learns, what is gradient

32
00:01:18,659 --> 00:01:21,360
descent, and what are activation

33
00:01:21,360 --> 00:01:26,729
functions. In module 3, we will learn

34
00:01:26,729 --> 00:01:28,590
about some of the most popular deep

35
00:01:28,590 --> 00:01:31,799
learning libraries, namely Keras, PyTorch,

36
00:01:31,799 --> 00:01:35,100
and TensorFlow. And we will learn

37
00:01:35,100 --> 00:01:37,320
how to use the Keras library to build

38
00:01:37,320 --> 00:01:39,049
models for regression and classification

39
00:01:39,049 --> 00:01:44,250
problems. In module 4, we will learn about

40
00:01:44,250 --> 00:01:46,860
supervised and unsupervised deep neural

41
00:01:46,860 --> 00:01:49,020
networks, namely convolutional neural

42
00:01:49,020 --> 00:01:52,040
networks, recurrent neural networks. and

43
00:01:52,040 --> 00:01:55,170
autoencoders. We will also learn how to

44
00:01:55,170 --> 00:01:57,420
use the Keras library to build a

45
00:01:57,420 --> 00:01:59,120
convolutional neural network.

46
00:02:02,840 --> 00:02:04,200
Before I conclude this video,

47
00:02:04,200 --> 00:02:06,299
I would like to make an important note.

48
00:02:06,299 --> 00:02:09,780
In this course, I decided to focus on the

49
00:02:09,780 --> 00:02:12,280
fundamentals of deep learning. Deep

50
00:02:12,280 --> 00:02:13,319
learning is a vast

51
00:02:13,319 --> 00:02:15,299
field now and is continuously evolving

52
00:02:15,299 --> 00:02:18,599
at a fast pace. Therefore, the field can

53
00:02:18,599 --> 00:02:21,719
be intimidating to many people, so I have

54
00:02:21,719 --> 00:02:23,730
tailored this course to those who really

55
00:02:23,730 --> 00:02:25,709
know nothing about deep learning or

56
00:02:25,709 --> 00:02:28,590
neural networks. If you have worked a lot

57
00:02:28,590 --> 00:02:30,840
with artificial neural networks and deep

58
00:02:30,840 --> 00:02:32,969
learning models, then maybe this course

59
00:02:32,969 --> 00:02:35,040
is not for you. You are more than welcome

60
00:02:35,040 --> 00:02:36,870
to still take it as a refresher,

61
00:02:36,870 --> 00:02:39,629
but maybe you will not learn a lot. While I

62
00:02:39,629 --> 00:02:41,519
will introduce you to some advanced

63
00:02:41,519 --> 00:02:43,739
topics in deep learning, but I am just

64
00:02:43,739 --> 00:02:45,480
going to give you a simplified version

65
00:02:45,480 --> 00:02:46,940
of them. I just wanted to

66
00:02:46,940 --> 00:02:49,580
make this point clear in order to set

67
00:02:49,580 --> 00:02:52,180
the expectations right at the beginning

68
00:02:52,180 --> 00:02:55,200
of the course. And now with that, welcome

69
00:02:55,200 --> 00:02:57,320
again and let's get started!