I am currently studying Kernel Ridge Regression. Specifically, I am considering the radial basis function kernel. Throughout literature I am seeing plots of the eigenfunctions that are the approximations of the eigenfunctions of the Kernel Operator. Below is a picture of the first 16 eigenfunctions from (Tibshirani et al 2008). I would like to know how to find an explicit expression for these functions. Perhaps there is a function in the R package, kernlab.

How can I represent the eigenvectors of kernel matrix on $\mathbb{R}$? Also, are these scaled eigenfunctions that are the feature space essentially the first 16 scores of Kernel Principal Component Analysis using the RBF?

from The Elements of Statistical Learning


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.