#### Preview

Consider the bounded linear operator, *L*: *F* → *Z*, where *Z* ⊆ R^{N} and F are Hilbert spaces defined on a common field *X*. *L* is made up of a series of *N* bounded linear evaluation functionals, *L*_{i}: *F* → R. By the Riesz representation theorem, there exist functions *k*(*x*_{i}, ˙) ∈ *F* : *L*_{i}*f* = 〈*f*, *k*(*x*_{i}, ˙)〉_{F}. The functions, *k*(*x*_{i}, ˙), are known as reproducing kernels and *F* is a reproducing kernel Hilbert space (RKHS). This is a natural framework for approximating functions given a discrete set of observations. In this paper the computational aspects of characterizing such approximations are described and a gradient method presented for iterative solution. Such iterative solutions are desirable when *N* is large and the matrix computations involved in the basic solution become infeasible. This is also exactly the case where the problem becomes ill-conditioned. An iterative approach to Tikhonov regularization is therefore also introduced. Unlike iterative solutions for the more general Hilbert space setting, the proofs presented make use of the spectral representation of the kernel.

*Keywords: *
reproducing kernel Hilbert space; gradient iteration; function approximation

*Journal Article.*
*0 words.*

*Subjects: *
Mathematics

Go to Oxford Journals » abstract

Full text: subscription required

How to subscribe Recommend to my Librarian

Users without a subscription are not able to see the full content. Please, subscribe or login to access all content.