Dodd, T.J. and Harrison, R.F. (2001) The method of Successive Approximations for Reproducing Kernel Hilbert Spaces (1). Research Report. ACSE Research Report 805 . Department of Automatic Control and Systems Engineering
Abstract
A general framework for function approximation from finite data is presented based on reproducing kernel Hilbert spaces. Key results are summarised and the normal and regularised solutions are described. A potential limitation to these solutions for large data sets is the computational burden. An iterative approach to the least squares normal solution is proposed to overcome this. Detailed proofs of convergence are given.
Metadata
Item Type: | Monograph |
---|---|
Authors/Creators: |
|
Dates: |
|
Institution: | The University of Sheffield |
Academic Units: | The University of Sheffield > Faculty of Engineering (Sheffield) > Department of Automatic Control and Systems Engineering (Sheffield) > ACSE Research Reports |
Depositing User: | MRS ALISON THERESA BARNETT |
Date Deposited: | 23 Jan 2015 11:36 |
Last Modified: | 25 Oct 2016 02:29 |
Status: | Published |
Publisher: | Department of Automatic Control and Systems Engineering |
Series Name: | ACSE Research Report 805 |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:83055 |