L1/2 regularization

被引：0
作者：
ZongBen Xu
Hai Zhang
Yao Wang
XiangYu Chang
Yong Liang
机构：
[1] Xi’an Jiaotong University,Institute of Information and System Science
[2] Northwest University,Department of Mathematics
[3] University of Science and Technology,undefined
来源：
Science China Information Sciences | 2010年 / 53卷
关键词：
machine learning; variable selection; regularizer; compressed sensing;
D O I：
暂无
中图分类号：
学科分类号：
摘要：
In this paper we propose an L1/2 regularizer which has a nonconvex penalty. The L1/2 regularizer is shown to have many promising properties such as unbiasedness, sparsity and oracle properties. A reweighed iterative algorithm is proposed so that the solution of the L1/2 regularizer can be solved through transforming it into the solution of a series of L1 regularizers. The solution of the L1/2 regularizer is more sparse than that of the L1 regularizer, while solving the L1/2 regularizer is much simpler than solving the L0 regularizer. The experiments show that the L1/2 regularizer is very useful and efficient, and can be taken as a representative of the Lp(0 > p > 1)regularizer.
引用
收藏
页码：1159 / 1169
页数：10
相关论文
共 43 条
  • [1] Schwarz G.(1978)Estimating the dimension of a model Ann Statist 6 461-464
  • [2] Mallows C. L.(1973)Some comments on Cp Technometrics 15 661-675
  • [3] Tibshirani R.(1996)Regression shrinkage and selection via the Lasso J Royal Statist Soc B 58 267-288
  • [4] Donoho D. L.(2001)Uncertainty principles and ideal atomic decomposition IEEE Trans Inf Theory 47 2845-2862
  • [5] Huo X.(2003)Maximal sparsity representation via Proc Natl Acal Sci 100 2197-2202
  • [6] Donoho D. L.(2001) minimization SIAM Rev 43 129-159
  • [7] Elad E.(2004)Atomic decomposition by basis pursuit Ann Statist 32 928-961
  • [8] Chen S.(2006)Nonconcave penalty likelihood with a diverging number of parameters J Amer Statist Assoc 101 1418-1429
  • [9] Donoho D. L.(2005)The adaptive Lasso and its oracle properties J Royal Statist Soc B 67 301-320
  • [10] Saunders M.(2007)Regularization and variable selection via the elastic net J Mach Learn Res 8 2701-2726
12345