Sciweavers

Share
NIPS
2004

Learning, Regularization and Ill-Posed Inverse Problems

9 years 24 days ago
Learning, Regularization and Ill-Posed Inverse Problems
Many works have shown that strong connections relate learning from examples to regularization techniques for ill-posed inverse problems. Nevertheless by now there was no formal evidence neither that learning from examples could be seen as an inverse problem nor that theoretical results in learning theory could be independently derived using tools from regularization theory. In this paper we provide a positive answer to both questions. Indeed, considering the square loss, we translate the learning problem in the language of regularization theory and show that consistency results and optimal regularization parameter choice can be derived by the discretization of the corresponding inverse problem.
Lorenzo Rosasco, Andrea Caponnetto, Ernesto De Vit
Added 31 Oct 2010
Updated 31 Oct 2010
Type Conference
Year 2004
Where NIPS
Authors Lorenzo Rosasco, Andrea Caponnetto, Ernesto De Vito, Francesca Odone, Umberto De Giovannini
Comments (0)
books