Asymptotic Statistical Analysis of Sparse Group LASSO via Approximate Message Passing Algorithm

Abstract

Sparse Group LASSO (SGL) is a regularized model for high-dimensional linear regression problems with grouped covariates. SGL applies $l_1$ and $l_2$ penalties on the individual predictors and group predictors, respectively, to guarantee sparse effects both on the inter-group and within-group levels. In this paper, we apply the approximate message passing (AMP) algorithm to efficiently solve the SGL problem under Gaussian random designs. We further use the recently developed state evolution analysis of AMP to derive an asymptotically exact characterization of SGL solution. This allows us to conduct multiple fine-grained statistical analyses of SGL, through which we investigate the effects of the group information and $\gamma$ (proportion of ${\ell }_1$ penalty). With the lens of various performance measures, we show that SGL with small $\gamma$ benefits significantly from the group information and can outperform other SGL (including LASSO) or regularized models which do not exploit the group information, in terms of the recovery rate of signal, false discovery rate and mean squared error.

References in corpus (5)

• Adaptive Damping and Mean Removal for the Generalized Approximate Message Passing Algorithm
• Near optimal compressed sensing without priors: Parametric SURE Approximate Message Passing
• A Primer on Coordinate Descent Algorithms
• Exact block-wise optimization in group lasso and sparse group lasso for linear regression
• Approximate message-passing for convex optimization with non-separable penalties