1
00:00:00,180 --> 00:00:05,120
This one, like the weirdest thing, I was like trying to figure, I assume there's gonna

2
00:00:05,120 --> 00:00:10,100
be a very elegant name for this but this is literally pronounced pass hat K. And it means

3
00:00:10,100 --> 00:00:15,440
if I run it K times, am I gonna get the right answer each of the K times? And that is

4
00:00:15,440 --> 00:00:20,980
the reliability one. And this becomes more important when the outcome is harder to measure,

5
00:00:21,480 --> 00:00:25,320
right? When it's more subjective, it's more open -ended and you can't really detect,

6
00:00:25,360 --> 00:00:30,040
then you want like the reliable model to be doing your bidding.

7
00:00:33,040 --> 00:00:39,380
Like I would not put a, like one example of this is like Mark mentioned, like there are

8
00:00:39,380 --> 00:00:44,060
things that can like land you in jail if you like break policies, right? Like trade stocks

9
00:00:44,060 --> 00:00:48,760
doing, when you're not supposed to, you don't wanna be experimenting with, you know.

10
00:00:48,880 --> 00:00:52,920
If it gets it right one out of 10 times, like nine out of 10 times, you're in jail. And

11
00:00:52,920 --> 00:00:56,840
you also don't wanna run the experiment 10 times because then you might go to jail nine

12
00:00:56,840 --> 00:00:57,500
times.