Re:Major research breakthroughs in the last 10-15 years
by henning on Thursday September 20, 2007 @02:45AM
(#57280)
Support vector machines and other practical realizations of statistical learning theory research that began in the 1970s.
[ Parent
| on HN ]
Re:Major research breakthroughs in the last 10-15 years
by viergroupie on Thursday September 20, 2007 @12:07PM
(#57445)
Why single out support vector machines? Sure, the kernel trick is neat but there's really nothing about them that stands out above other classification methods. It would make more sense just to say "the field of machine learning".
Re:Major research breakthroughs in the last 10-15 years
by henning on Thursday September 20, 2007 @03:25PM
(#57545)
It's a practical (especially for small to medium sized data sets), effective way of doing machine learning that has a highly sound theory behind it and doesn't require arcane tricks that are rarely or never written down, unlike old-style artificial neural networks.
The interest in kernel methods is proof that there's nothing more practical than a good theory. :)
Re:Major research breakthroughs in the last 10-15 years
by viergroupie on Friday September 21, 2007 @12:52AM
(#57819)
At the same time, I think it's astounding how often k-nearest neighbors (a method considerably simpler than SVMs) is either the best or second-best classifier for a dataset. I'm not trying to pit SVMs vs. neural networks...rather I'm saying most methods of machine learning (neural nets, SVMs, gaussian mixture, kNN, what have you) can be effective and none stand out above the rest.
Re:Major research breakthroughs in the last 10-15 years
by viergroupie on Friday September 21, 2007 @12:42PM
(#57994)
Woops...forgot the squeeze an "often" before the "either"...