Title: "A Meta-Learning Approach to the Regularized Learning - Case Study: Blood Glucose Prediction".

Abstract: We are going to present a new scheme of a regularization
kernel based learning method, where the kernel and the regularization
parameter are adaptively chosen within the regularization procedure.
The construction of such fully adaptive regularization algorithm is
motivated by the problem of predicting the blood glucose concentration
of diabetic patients. We describe how proposed scheme can be used for
this purpose and report the results of numerical experiments with real
clinical data.
The presentation is based on the joint research with Valeriya Naumova (RICAM),
Sivananthan Sampath (RICAM), Jette Randlov (Novo Nordisk A/S) and Samuel McKennoch
(Novo Nordisk A/S).