demon

demon is a function that implements GSparO for an example of least squares regression with A and b being Gaussian ensembles. A figure plotting the true signal and estimation by GSparO is illustrated in Plots, and the errors of least squares regression and obtained solution are printed. Two packages ThreeWay and ggplot2 should be installed for implementing demon.

Approaches a group sparse solution of an underdetermined linear system. It implements the proximal gradient algorithm to solve a lower regularization model of group sparse learning. For details, please refer to the paper "Y. Hu, C. Li, K. Meng, J. Qin and X. Yang. Group sparse optimization via l_{p,q} regularization. Journal of Machine Learning Research, to appear, 2017".

demon: The example for GSparO

Description

Usage

Arguments

Details

Examples