Consider a positive-definite, symmetric function $k(x_1, x_2)$ which is used, given the dataset $\{(x_i, y_i)\}_{i=1}^m$, to construct the Gram matrix $K = [k(x_i, x_j)]_{i,j \in 1, ..., m}$.
What is the relation between generalization bounds of Kernel Ridge Regression using $K$ to solve the supervised learning problem and the largest eigenvalue of the $K$? Can you give a reference?
