1
00:00:00,211 --> 00:00:04,140
Hello everyone, and welcome to the class on probabilistic graphical models.

2
00:00:04,140 --> 00:00:07,647
My name is Daphne Koller and I’m a professor at Stanford University.

3
00:00:07,647 --> 00:00:09,618
We here at Stanford are really excited

4
00:00:09,618 --> 00:00:12,209
to be able to offer this graduate level Stanford class

5
00:00:12,209 --> 00:00:14,751
to anyone, anywhere around the world for free.

6
00:00:14,751 --> 00:00:18,481
So what are probabilistic graphical models?

7
00:00:18,481 --> 00:00:20,751
Well, it’s a bit complicated to explain

8
00:00:20,751 --> 00:00:24,095
and we’re going to talk about that in an upcoming video

9
00:00:24,095 --> 00:00:26,192
but also throughout the entire class.

10
00:00:26,192 --> 00:00:29,716
In this video, I’d like to tell you a little bit about the format of this class.

11
00:00:29,716 --> 00:00:32,816
The course is going to be offered over ten weeks worth of material

12
00:00:32,816 --> 00:00:34,794
plus a final examination at the end.

13
00:00:34,794 --> 00:00:39,066
The content is going to be conveyed via a set of videos,

14
00:00:39,066 --> 00:00:42,317
augmented with quizzes to reinforce understanding.

15
00:00:42,317 --> 00:00:45,496
In addition, there is going to be a weekly problem set

16
00:00:45,496 --> 00:00:50,041
where the problem sets altogether are going to be worth 25% of the score

17
00:00:50,041 --> 00:00:53,479
for a total of the nine problem sets for the nine weeks worth of content.

18
00:00:53,479 --> 00:00:56,568
The problem sets are designed to allow for multiple submissions,

19
00:00:56,568 --> 00:01:00,626
so that each version of the problem set is going to be a little bit different

20
00:01:00,626 --> 00:01:04,040
so that you can resubmit the same problem set [a] couple of times

21
00:01:04,040 --> 00:01:06,623
to make sure that you really mastered the material.

22
00:01:06,623 --> 00:01:10,962
In addition, there’s going to be a weekly programming assignment,

23
00:01:10,962 --> 00:01:13,859
and those programming assignments were selected

24
00:01:13,859 --> 00:01:17,976
to reinforce specific concepts that we’re studying in the course,

25
00:01:17,976 --> 00:01:21,343
but at the same time to reveal the range of applications

26
00:01:21,343 --> 00:01:25,431
to which the framework of probabilistic graphical models can be successfully applied.

27
00:01:25,431 --> 00:01:27,288
So we’re going to have, for example,

28
00:01:27,288 --> 00:01:30,367
a problem set on how you use probabilistic graphical models

29
00:01:30,367 --> 00:01:33,639
to understand the inheritance of genetically inherited diseases.

30
00:01:33,639 --> 00:01:35,647
We’re going to have one that shows

31
00:01:35,647 --> 00:01:38,963
how you can look at a set of handwritten characters

32
00:01:38,963 --> 00:01:40,937
and read what’s written there.

33
00:01:40,937 --> 00:01:44,087
And we’re going to have one that allows you

34
00:01:44,087 --> 00:01:47,647
to look at a stream of output from a Kinect sensor

35
00:01:47,647 --> 00:01:49,759
that gives you both video and range data

36
00:01:49,759 --> 00:01:52,167
and recognize human activities.

37
00:01:52,167 --> 00:01:56,458
These nine programming assignments are each going to be worth 7% of the score

38
00:01:56,458 --> 00:01:58,089
for a total of 63%,

39
00:01:58,089 --> 00:02:01,408
which gives us 12% left for the final exam.

40
00:02:01,408 --> 00:02:04,480
What background do you need for this class?

41
00:02:04,480 --> 00:02:07,540
Well, it’s going to be really hard to do this class,

42
00:02:07,540 --> 00:02:11,048
without some understanding of basic probability theory.

43
00:02:11,048 --> 00:02:13,087
This doesn’t have to be very advanced stuff.

44
00:02:13,087 --> 00:02:16,176
We’re talking about things like independence and Bayes' rule

45
00:02:16,176 --> 00:02:18,711
And just basics of discrete distributions.

46
00:02:18,711 --> 00:02:20,615
And we also have a few introductory modules

47
00:02:20,615 --> 00:02:23,687
to refresh your memory about these basic concepts.

48
00:02:23,687 --> 00:02:26,942
The programming assignments will require

49
00:02:26,942 --> 00:02:29,416
that you’ve had some experience programming before

50
00:02:29,416 --> 00:02:30,871
because this is not a programming class.

51
00:02:30,871 --> 00:02:31,959
We don’t teach you how to program.

52
00:02:31,959 --> 00:02:37,383
And because this class merges ideas from both probability theory and computer science,

53
00:02:37,383 --> 00:02:41,247
it’s really important you have some background in algorithms and data structures.

54
00:02:41,247 --> 00:02:44,475
Recommended, but not strictly necessary—

55
00:02:44,475 --> 00:02:46,597
and we certainly don’t require it,

56
00:02:46,597 --> 00:02:48,719
and we give you the background as we go—

57
00:02:48,719 --> 00:02:52,290
is a little bit of experience, perhaps, in machine learning,

58
00:02:52,290 --> 00:02:55,367
maybe some simple optimization like gradient descent,

59
00:02:55,367 --> 00:02:57,445
nothing very sophisticated.

60
00:02:57,445 --> 00:03:01,483
And it would be helpful to have some experience programming in Matlab or Octave,

61
00:03:01,483 --> 00:03:04,364
although, here also, we have some introductory modules

62
00:03:04,364 --> 00:03:08,337
that help you learn this programming language if you haven’t played around with it before.

63
00:03:10,044 --> 00:03:12,482
A few other issues that are worth noting:

64
00:03:12,482 --> 00:03:14,552
This class has an honor code.

65
00:03:14,552 --> 00:03:19,017
This is the norm also for our local Stanford students when they’re taking a Stanford class.

66
00:03:19,017 --> 00:03:22,836
The honor code here says that you’re allowed to discuss the material,

67
00:03:22,836 --> 00:03:26,583
in fact even encouraged to discuss the material with your fellow classmates.

68
00:03:26,583 --> 00:03:30,809
You can even ask clarifying questions about the problems sets and the programming assignments.

69
00:03:30,809 --> 00:03:32,991
But what you turn in has to be your own work.

70
00:03:32,991 --> 00:03:39,537
Furthermore, we request that you do not post either the programming assignments

71
00:03:39,537 --> 00:03:41,711
or their solutions anywhere on the web,

72
00:03:41,711 --> 00:03:44,601
so that future generations of students can do

73
00:03:44,601 --> 00:03:48,447
the problems sets and the programming assignments independently as well.

74
00:03:48,447 --> 00:03:52,359
A second issue to keep in mind is that of time management.

75
00:03:52,359 --> 00:03:54,375
This is a graduate-level Stanford class

76
00:03:54,375 --> 00:03:56,863
and it’s considered a difficult one even at Stanford.

77
00:03:56,863 --> 00:03:59,146
A typical Stanford student can easily spend

78
00:03:59,146 --> 00:04:01,073
ten to fifteen hours a week on this class,

79
00:04:01,073 --> 00:04:02,904
and so we would suggest that you budget

80
00:04:02,904 --> 00:04:05,791
at least that amount of time for your own efforts on this class

81
00:04:05,791 --> 00:04:08,299
if you don’t want to find yourself running out of time

82
00:04:08,299 --> 00:04:10,375
when a submission deadline comes around.

83
00:04:10,375 --> 00:04:13,341
We’ve built in a little slack into the submission deadline,

84
00:04:13,341 --> 00:04:17,655
so if you don’t manage to submit by the original deadline,

85
00:04:17,655 --> 00:04:19,362
you have a week’s worth of grace period.

86
00:04:19,362 --> 00:04:22,975
But then that obviously starts to impinge on the next week’s problem set.

87
00:04:22,975 --> 00:04:25,439
So we advise that you don’t just keep

88
00:04:25,439 --> 00:04:28,511
a backlog of assignments throughout the course,

89
00:04:28,511 --> 00:04:31,482
because it will all end up coming back to bite you in the end.

90
00:04:32,021 --> 00:04:36,347
Finally, part of the experience of this class

91
00:04:36,347 --> 00:04:38,591
is interacting with your fellow students,

92
00:04:38,591 --> 00:04:41,447
so for that purpose we have the discussion forum

93
00:04:41,447 --> 00:04:43,269
which has proven in other classes

94
00:04:43,269 --> 00:04:46,527
to be an invaluable resource for interacting with other students,

95
00:04:46,527 --> 00:04:49,783
asking questions and obtaining a deeper understanding of the material.

96
00:04:49,783 --> 00:04:52,412
We’re also encouraging you to form study groups—

97
00:04:52,412 --> 00:04:55,892
these can be physical study groups with people in the same geographical region,

98
00:04:55,892 --> 00:05:00,280
or online study groups where you can just discuss the material with each other.

99
00:05:00,280 --> 00:05:01,839
We believe that doing this

100
00:05:01,839 --> 00:05:03,933
will give you a much better understanding of the material

101
00:05:03,933 --> 00:05:06,832
and will make the course considerably more fun as well.

102
00:05:06,832 --> 00:05:09,368
So, to summarize,

103
00:05:09,368 --> 00:05:14,999
through all these different pieces of the content and the exercises,

104
00:05:14,999 --> 00:05:17,831
we think that you’ll learn fundamental methods

105
00:05:17,831 --> 00:05:19,880
in this area of probabilistic graphical models.

106
00:05:19,880 --> 00:05:22,342
You’ll also get to see and play around with

107
00:05:22,342 --> 00:05:25,872
a range of real-world applications for which these methods have been applied

108
00:05:25,872 --> 00:05:29,346
and hopefully you will leave this class with an understanding

109
00:05:29,346 --> 00:05:32,314
of how to take these ideas and use them in your own work

110
00:05:32,314 --> 00:05:33,875
in problems that you care about.

111
00:05:33,875 --> 99:59:59,000
We look forward to seeing you in this class.