Kernel ridge regression. (English) Zbl 1325.62094

Vovk, Vladimir

Schölkopf, Bernhard (ed.) et al., Empirical inference. Festschrift in honor of Vladimir N. Vapnik. Berlin: Springer (ISBN 978-3-642-41135-9/hbk; 978-3-642-41136-6/ebook). 105-116 (2013).

Summary: This chapter discusses the method of Kernel Ridge Regression, which is a very simple special case of Support Vector Regression. The main formula of the method is identical to a formula in Bayesian statistics, but Kernel Ridge Regression has performance guarantees that have nothing to do with Bayesian assumptions. I will discuss two kinds of such performance guarantees: those not requiring any assumptions whatsoever, and those depending on the assumption of randomness.
For the entire collection see [Zbl 1280.68040].

MSC:

62G08	Nonparametric regression and quantile regression
62J07	Ridge regression; shrinkage estimators (Lasso)
62H30	Classification and discrimination; cluster analysis (statistical aspects)