Convergence analysis of online algorithms for vector-valued kernel regression

Michael Griebel, Peter Oswald

We consider the problem of approximating the regression function $f_{\hat{I}}¼:\hat{I}\text{c◯}\to Y$ from noisy $\hat{I}¼$-distributed vector-valued data $(\ddot{I}{‰}_m,y_m)\in \hat{I}\text{c◯}\times Y$ by an online learning al…