LFD Book Forum

LFD Book Forum (http://book.caltech.edu/bookforum/index.php)
-   The Final (http://book.caltech.edu/bookforum/forumdisplay.php?f=138)
-   -   *answer* q2 (http://book.caltech.edu/bookforum/showthread.php?t=4624)

allenyin 09-28-2015 07:09 PM

*answer* q2
 
Can anyone explain to me why the answer to this question is logistic regression?

Logistic regression models trained on different data-sets will all yield value in the range [0,1] given new test points. Averaging [0,1] will still result in [0,1], wouldn't that also make g_avg a valid logistic regression?

The only thing I can think of might be that, the average might end up with total probability integrated in the interval to be greater than 1...

yaser 09-28-2015 09:53 PM

Re: *answer* q2
 
The key point is whether the average can be written as a single logistic regression with the appropriate parameters. If it can, then \bar g \in {\cal H} since {\cal H} contains all logistic regression functions. If sometimes it cannot, then indeed \bar g may not be in {\cal H}. Can you think of two logistics regression functions whose average is not one?


All times are GMT -7. The time now is 02:24 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.