A methodology to perform topological regularization via information filtering network is introduced. This methodology can be directly applied to sparse modeling with the vast family of elliptical probability distributions. It can also be directly implemented for $L_0$ norm regularized multicollinear regression. In this paper, I describe in detail an application to sparse modeling with multivariate Student-t. A specific $L_0$ norm regularized expectation-maximization likelihood maximization procedure is proposed for this sparse Student-t case. Examples with real data from stock prices log-returns and from artificially generated data demonstrate applicability, performances, and potentials of this methodology.