1: \begin{abstract}

2: We study generalization properties of kernel regularized least squares regression based on a partitioning approach. We show that 

3: optimal rates of convergence are preserved  if the number of local sets grows sufficiently slowly with the sample size. 

4: Moreover, the partitioning approach can be efficiently combined with local Nystr\"om subsampling, improving computational cost twofold. 

5: \end{abstract}