Liu Bie Ju Centre for Mathematical Sciences
City University of Hong Kong
Mathematical Analysis and its Applications
Colloquium

Organized by Prof. Philippe G. Ciarlet and Prof. Roderick Wong

Properties of Regularization Operators in Learning Theory

by
Dr. Andrea Caponnetto
Department of Mathematics
City University of Hong Kong

Date: October 17, 2007 (Wednesday)
Time: 4:30 pm to 5:30 pm
Venue: Room B6605 (Faculty Conference Room)
Blue Zone, Level 6
Academic Building
City University of Hong Kong

Abstract: We discuss the properties of a large class of learning algorithms defined in terms of classical regularization operators for ill-posed problems. This class includes regularized least-squares, Landweber method and truncated singular value decomposition over hypothesis spaces defined as reproducing kernel Hilbert spaces of vector-valued functions.

A minimax analysis for convergence rates over suitable priors will be presented, together with data-dependent choices of the regularization parameter enforcing statistical adaptation.
             

** All interested are welcome **

For enquiry: 2788-9816