Home Research
 
Home Research

Regularized kernel methods

Research Area: Learning Theory and Algorithms
Status: In progress  
Faculty: Alessandro Verri, Ernesto De Vito Participants: Paolo Albini, Veronica Umanità, Lorenzo Rosasco
 
Description:

The starting point of our work is the connection between learning theory and the theory of regularization of ill-posed inverse problems. Our goals are:

  • to study consistency and finite sample bounds of regularized learning algorithms
  • to design and develop learning algorithms based on spectral filters
  • to investigate the problems of vector valued regression