LFD Book Forum

LFD Book Forum (http://book.caltech.edu/bookforum/index.php)
-   Chapter 4 - Overfitting (http://book.caltech.edu/bookforum/forumdisplay.php?f=111)
-   -   Data snooping (http://book.caltech.edu/bookforum/showthread.php?t=4458)

cma61 11-09-2013 01:56 PM

Data snooping
 
I still cannot understand the data snooping idea. Does that mean we can't look at the data before we choose the hypothesis set? What about Fig 3.6 in the book?

yaser 11-17-2013 03:29 AM

Re: Data snooping
 
Quote:

Originally Posted by cma61 (Post 11604)
I still cannot understand the data snooping idea. Does that mean we can't look at the data before we choose the hypothesis set? What about Fig 3.6 in the book?

Sorry for the delay as I am attending to the edX forum this term.

You cannot look at the data and expect the generalization bound you get for the model that you chose based on what you have seen to be valid.

Figure 3.6 is just illustrative of the nonlinear transform process, without worrying about the generalization issues.


All times are GMT -7. The time now is 10:05 PM.

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.