Sparse recovery in convex hulls via entropy penalization

Let be a random couple in with unknown distribution and be i.i.d. copies of Denote the empirical distribution of Let be a dictionary that consists of functions. For denote Let be a given loss function and suppose it is convex with respect to the second variable. Let Finally, let be the simplex of all probability distributions on Consider the following penalized empirical risk minimization problem \begin{eqnarray*}\hat{\lambda}^{\varepsilon}:={\mathop {argmin}_{\lambda\in \Lambda}}\Biggl[P_n(\ell \bullet f_{\lambda})+\varepsilon \sum_{j=1}^N\lambda_j\log \lambda_j\Biggr]\end{eqnarray*} along with its distribution dependent version \begin{eqnarray*}\lambda^{\varepsilon}:={\mathop {argmin}_{\lambda\in \Lambda}}\Biggl[P(\ell \bullet f_{\lambda})+\varepsilon \sum_{j=1}^N\lambda_j\log \lambda_j\Biggr],\end{eqnarray*} where is a regularization parameter. It is proved that the ``approximate sparsity'' of implies the ``approximate sparsity'' of and the impact of ``sparsity'' on bounding the excess risk of the empirical solution is explored. Similar results are also discussed in the case of entropy penalized density estimation.
View on arXiv