LFD Book Forum

LFD Book Forum (http://book.caltech.edu/bookforum/index.php)
-   Homework 8 (http://book.caltech.edu/bookforum/forumdisplay.php?f=137)
-   -   Learning performance comparable to SVM that doesnt require QP (http://book.caltech.edu/bookforum/showthread.php?t=4390)

hsolo 07-23-2013 06:05 PM

Learning performance comparable to SVM that doesnt require QP
 
1. Is there any learning algorithm/approach that offers, in practice, performance comparable to SVM but that doesnt require QP?

For eg, http://cbcl.mit.edu/cbcl/publications/ps/rlsc.pdf

2. At a high level is it correct to think of regularization as introducing a 'softness' and thus a 'generalization dividend' for regression problems and of SVMs (soft kernels) as introducing a generalization dividend for classification problems, albeit at the cost of QP?

yaser 07-24-2013 08:48 PM

Re: Learning performance comparable to SVM that doesnt require QP
 
Quote:

Originally Posted by hsolo (Post 11298)
1. Is there any learning algorithm/approach that offers, in practice, performance comparable to SVM but that doesnt require QP?

For eg, http://cbcl.mit.edu/cbcl/publications/ps/rlsc.pdf

2. At a high level is it correct to think of regularization as introducing a 'softness' and thus a 'generalization dividend' for regression problems and of SVMs (soft kernels) as introducing a generalization dividend for classification problems, albeit at the cost of QP?

1. Different techniques outperform others in different problems, so other methods, including the one you mention, will indeed beat SVM in some cases. None of the methods wins in all applications.

2. This is a legitimate way of looking at it.


All times are GMT -7. The time now is 03:52 AM.

Powered by vBulletin® Version 3.8.3
Copyright ©2000 - 2020, 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.