LFD Book Forum

LFD Book Forum (http://book.caltech.edu/bookforum/index.php)
-   Chapter 1 - The Learning Problem (http://book.caltech.edu/bookforum/forumdisplay.php?f=108)
-   -   Hoeffding Inequality With Probability > 1? (http://book.caltech.edu/bookforum/showthread.php?t=4675)

henry2015 05-21-2016 12:29 AM
Hoeffding Inequality With Probability > 1?
 
Given that the inequality shows an upper bound of 2*e^(-2*N*(epsilon)^2).

If we have N = 10, and epsilon = 0.1, then we have

2*e^(-2*10*(0.1^2)) = 1.637461506.

So what does it mean? I thought we would never have a probability > 1.

ntvy95 05-21-2016 06:46 AM
Re: Hoeffding Inequality With Probability > 1?
 
Quote:
Originally Posted by henry2015 (Post 12363)
Given that the inequality shows an upper bound of 2*e^(-2*N*(epsilon)^2).

If we have N = 10, and epsilon = 0.1, then we have

2*e^(-2*10*(0.1^2)) = 1.637461506.

So what does it mean? I thought we would never have a probability > 1.
I think you can check out the first question in the Q&A section of Lecture 02.

henry2015 05-21-2016 01:03 PM
Re: Hoeffding Inequality With Probability > 1?
 
Quote:
Originally Posted by ntvy95 (Post 12366)
I think you can check out the first question in the Q&A section of Lecture 02.
Thanks for the link!


All times are GMT -7. The time now is 07:02 PM.

Powered by vBulletin® Version 3.8.3
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.
The contents of this forum are to be used ONLY by readers of the Learning From Data book by Yaser S. Abu-Mostafa, Malik Magdon-Ismail, and Hsuan-Tien Lin, and participants in the Learning From Data MOOC by Yaser S. Abu-Mostafa. No part of these contents is to be communicated or made accessible to ANY other person or entity.