PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Non linear inverse problem: complexity regularization
Jean-Michel Loubes and Carenne Ludena
(2004) Technical Report. Université Paris SUd, Paris.

Abstract

We provide oracle inequalities for non linear inverse problems. We build estimators based on model selection estimators and giver their rate of convergence. We prove, that provided good assumptions on the operator, non linearity does not change the rate of convergence and can be handled.

EPrint Type:Monograph (Technical Report)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Computational, Information-Theoretic Learning with Statistics
Learning/Statistics & Optimisation
ID Code:621
Deposited By:Jean-Michel Loubes
Deposited On:29 December 2004