1
00:00:00,290 --> 00:00:01,350
HI, I'm a Mac.

2
00:00:01,350 --> 00:00:03,780
>> [LAUGH] No. >> All right, we'll start over.

3
00:00:03,780 --> 00:00:04,600
Let's do this right.

4
00:00:04,600 --> 00:00:06,200
Hi, I'm Charles Isbeil.

5
00:00:06,200 --> 00:00:07,100
>> True.

6
00:00:07,100 --> 00:00:08,450
>> I'm a professor at Georgia Tech.

7
00:00:08,450 --> 00:00:09,130
>> True.

8
00:00:09,130 --> 00:00:10,520
>> This is Michael Littman.

9
00:00:10,520 --> 00:00:13,060
>> True. >> He is also a professor at Georgia Tech.

10
00:00:13,060 --> 00:00:14,040
>> False.

11
00:00:14,040 --> 00:00:16,510
>> No, that's actually true, while you're a professor at Brown,

12
00:00:16,510 --> 00:00:18,240
you're also an adjunct professor here at Georgia Tech.

13
00:00:20,150 --> 00:00:21,590
I'm a professional donut thief.

14
00:00:21,590 --> 00:00:22,190
>> False.

15
00:00:22,190 --> 00:00:24,020
>> That is correct I am not a professional.

16
00:00:24,020 --> 00:00:26,110
Together, we are teaching the course Machine Learning.

17
00:00:26,110 --> 00:00:27,000
>> True.

18
00:00:27,000 --> 00:00:30,070
>> The lectures will only be available on prime number calendar days.

19
00:00:30,070 --> 00:00:31,190
>> False.

20
00:00:31,190 --> 00:00:33,140
>> The mini course is on supervised learning.

21
00:00:33,140 --> 00:00:34,010
>> True.

22
00:00:34,010 --> 00:00:37,750
>> Supervised learning itself was invented in the 1830s by Gauss.

23
00:00:37,750 --> 00:00:38,710
>> False?

24
00:00:38,710 --> 00:00:39,450
>> Actually, I don't really know.

25
00:00:39,450 --> 00:00:41,680
It sounds like the sort of thing you'd invent but, probably not,

26
00:00:41,680 --> 00:00:43,250
so let's go with that.

27
00:00:43,250 --> 00:00:46,320
Okay, supervised learning is the problem of learning to map inputs to

28
00:00:46,320 --> 00:00:48,500
predictions of true or false.

29
00:00:48,500 --> 00:00:49,420
>> False.

30
00:00:49,420 --> 00:00:52,040
>> Good one, it also includes other kinds of predictions such as regression,

31
00:00:52,040 --> 00:00:54,190
where the output might be vectors or numbers.

32
00:00:54,190 --> 00:00:54,990
>> True.

33
00:00:54,990 --> 00:00:57,950
>> It's an important component of all sorts of technologies from stopping credit

34
00:00:57,950 --> 00:00:58,540
card fraud.

35
00:00:58,540 --> 00:01:00,660
>> True. >> To finding faces in camera images.

36
00:01:00,660 --> 00:01:03,070
>> True. >> To making taste your breath mints.

37
00:01:03,070 --> 00:01:04,050
>> False?

38
00:01:04,050 --> 00:01:07,232
>> I have no idea, but that sounded ridiculous so let's say false.

39
00:01:07,232 --> 00:01:08,350
To recognizing spoken language.

40
00:01:08,350 --> 00:01:09,220
>> True.

41
00:01:09,220 --> 00:01:12,090
>> Our goal here is the give you the skills that you need to recognize how to

42
00:01:12,090 --> 00:01:13,280
apply these technologies.

43
00:01:13,280 --> 00:01:13,910
>> True.

44
00:01:13,910 --> 00:01:15,340
>> And for interpreting their output, so

45
00:01:15,340 --> 00:01:17,300
that you can solve a range of data science problems.

46
00:01:17,300 --> 00:01:18,070
>> True.

47
00:01:18,070 --> 00:01:19,670
>> And for surviving a robot uprising.

48
00:01:19,670 --> 00:01:20,830
>> False. >> No, that is absolutely true.

49
00:01:20,830 --> 00:01:24,659
The most important thing you could get out of this course is learning how to

50
00:01:24,659 --> 00:01:26,712
survive the upcoming robot uprising.

51
00:01:26,712 --> 00:01:28,880
Okay, very good.

52
00:01:28,880 --> 00:01:31,062
You got an accuracy of 85%.

53
00:01:31,062 --> 00:01:31,767
>> True.

54
00:01:31,767 --> 00:01:34,873
>> Wrong, though after you answer this question,

55
00:01:34,873 --> 00:01:37,229
you will have an accuracy of 85%.

56
00:01:39,699 --> 00:01:40,590
>> True.

57
00:01:40,590 --> 00:01:43,070
>> That is correct, but false also would have been correct.

58
00:01:43,070 --> 00:01:43,882
Congratulations.