Description: In kernel ridge regression we have seen the final solution was not sparse in the variables ® .
We will now formulate a regression method that is sparse, i.e. it has the concept of support
vectors that determine the solution.
The thing to notice is that the sparseness arose from complementary slackness conditions
which in turn came from the fact that we had inequality constraints. In the SVM the penalty
that was paid for being on the wrong side of the support plane was given by C
P
i » k
i for
positive integers k, where » i is the orthogonal distance away from the support plane. Note
that the term jjwjj2 was there to penalize large w and hence to regularize the solution.
Importantly, there was no penalt
- [KPCAandSVM] - KPCA and SVM for face recognition SVM to
- [kerneladatron] - Kernel adatron, solving svm with gradien
- [SVM] - In this paper, we show how support vecto
- [KPLS] - Kernel-based partial least-squares algor
- [rbfSVM] - SVM algorithm based on RBF kernel
File list (Check if you may need any files):
SVregression.pdf
Download