skip to content
 

Model selection and sharp asymptotic minimaxity

Date: 
Tuesday 29th January 2008 - 11:00 to 12:30
Venue: 
INI Seminar Room 2
Abstract: 

We will show that a class of model selection procedures are asymptotically sharp minimax to recover sparse signals over a wide range of parameter spaces. Connections to Bayesian model selection, the MDL principle and wavelet estimation will be discussed.

Presentation Material: 
University of Cambridge Research Councils UK
    Clay Mathematics Institute London Mathematical Society NM Rothschild and Sons