Outlier-robust estimation of a sparse linear model
       using `1-penalized Huber’s M -estimator

                          Abstract
1      We study the problem of estimating a p-dimensional s-sparse vector in a linear
2      model with Gaussian design and additive noise. In the case where the labels are
3      contaminated by at most o adversarial outliers, we prove that the `1 -penalized
4      Huber’s M -estimator based on n samples attains the optimal rate of convergence
5      (s/n)1/ ...