Model Selection Consistency of Lasso for Empirical Data

Abstract

Large-scale empirical data, the sample size and the dimension are high, often exhibit various characteristics. For example, the noise term follows unknown distributions or the model is very sparse that the number of critical variables is fixed while dimensionality grows with n. The authors consider the model selection problem of lasso for this kind of data. The authors investigate both theoretical guarantees and simulations, and show that the lasso is robust for various kinds of data.