LFD Book Forum (http://book.caltech.edu/bookforum/index.php)
-   The Final (http://book.caltech.edu/bookforum/forumdisplay.php?f=138)
-   -   Clarification Request On Problem 13 (http://book.caltech.edu/bookforum/showthread.php?t=1503)

 munchkin 09-12-2012 08:04 PM

Clarification Request On Problem 13

If a data set is not separable using RBF in the Z-space then the QP solver will hit the iteration limit and bomb. If the solver completes and returns an alpha vector that happens to provide zero in-sample error with the test data set then how does that indicate that the training data was not linearly separable? Is every occurrence of zero in-sample error during testing to be interpreted as a failure to separate the training data? I don't understand the operating procedure specified by the problem text.

 yaser 09-12-2012 08:27 PM

Re: Clarification Request On Problem 14

Quote:
 Originally Posted by munchkin (Post 5203) If a data set is not separable using RBF in the Z-space then the QP solver will hit the iteration limit and bomb. If the solver completes and returns an alpha vector that happens to provide zero in-sample error with the test data set then how does that indicate that the training data was not linearly separable? Is every occurrence of zero in-sample error during testing to be interpreted as a failure to separate the training data? I don't understand the operating procedure specified by the problem text. Thanks for your attention.
If you get 's that achieve zero in-sample error, then the training data is separable by definition (those separating 's are an existence proof of that fact).

 munchkin 09-12-2012 09:44 PM

Re: Clarification Request On Problem 14

Thanks for the prompt response.

Ein of zero is perfectly separated. I understand that. Perhaps my confusion arose from thinking about the PLA and how it can fail spectacularly when confronted with non-separable data. I had the impression that missing a few target function points during the in-sample error calculation wasn't a problem of the same magnitude as the solver being unable to converge to any solution at all.

I will talley up the Ein=0 statistics as specified in the problem instructions. Thanks.

 All times are GMT -7. The time now is 07:47 AM.