

Structural Learning Seminar
November 18, 2016 17:00–19:00, Moscow, IITP, Bol'shoi Karetnyi per. 19 1






Threshold estimation for sparse highdimensional deconvolution
D. V. Belomestny^{} 
Number of views: 
This page:  35 

Abstract:
The problem of covariance estimation for a pdimensional normal vector X ∼ N(0, Σ) observed with additional noise is studied. Only a very general nonparametric assumption is imposed on the distribution of the noise. In this semiparametric deconvolution problem spectral thresholding estimators are constructed that adapt to sparsity in Σ. We prove that the minimax convergence rates logarithmic in log p/n with n being the sample size.

