Reducing training time by efficient localized kernel regression
Nicole Muecke · International Conference on Artificial Intelligence and Statistics · 2019
We study generalization properties of kernel regularized least squares regression based on a partitioning approach. We show that optimal rates of convergence are preserved if the number of local sets grows sufficiently slowly with the sample size. Moreover, the partitioning approach can be efficiently combined with local Nystrom subsampling, improving computational cost twofold.