Restricted eigenvalue property for corrupted Gaussian designs

05/21/2018

∙

by   Philip Thompson, et al.

∙

0

∙

share

Motivated by the construction of robust estimators using the convex relaxation paradigm, known to be computationally efficient, we present some conditions on the sample size which guarantee an augmented notion of Restricted Eigenvalue-type condition for Gaussian designs. Such notion is suitable for the construction of robust estimators of a multivariate Gaussian model whose samples are corrupted by outliers.