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)
-   -   Can we guarantee there is a function with low in sample error in H (http://book.caltech.edu/bookforum/showthread.php?t=4677)

pouramini 05-28-2016 06:18 AM

Can we guarantee there is a function with low in sample error in H
 
There is this sentence in book, page 25
Quote:

We cannot guarantee that we will find a hypothesis that achieves Ein (g) ~ 0,
but at least we will know if we find it
My English is not perfect, what does it mean? does it mean that we only should find such a function, otherwise there is no way to know if it exists?!

htlin 05-28-2016 09:49 AM

Re: Can we guarantee there is a function with low in sample error in H
 
Whether there is some g \in \mathcal{H} with E_{in} \sim 0 depends on the hypothesis set. So we cannot always guarantee finding a g with low E_{in}.

pouramini 05-29-2016 02:15 PM

Re: Can we guarantee there is a function with low in sample error in H
 
Thank you!:)


All times are GMT -7. The time now is 07:15 AM.

Powered by vBulletin® Version 3.8.3
Copyright ©2000 - 2019, 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.