Optimal learning of bandlimited functions from localized sampling (Q1023396): Difference between revisions

The core of this compactly written paper deals with the problem of learning a function in translation-invariant reproducing kernel Hilbert spaces. The main application is in the last section of the paper: an optimal algorithm for approximating bandlimited functions from localized sampling is established. In the univariate case upper and lower bound estimates for the approximation error are given.