期刊
SCIENCE CHINA-INFORMATION SCIENCES
卷 53, 期 6, 页码 1159-1169出版社
SCIENCE PRESS
DOI: 10.1007/s11432-010-0090-0
关键词
machine learning; variable selection; regularizer; compressed sensing
资金
- National Basic Research Program of China [2007CB311002]
- National Natural Science Foundation of China [60975036]
- Scientific Research Plan Projects of Shaanxi Education Department [08jk473]
In this paper we propose an L (1/2) regularizer which has a nonconvex penalty. The L (1/2) 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 L (1/2) regularizer can be solved through transforming it into the solution of a series of L (1) regularizers. The solution of the L (1/2) regularizer is more sparse than that of the L (1) regularizer, while solving the L (1/2) regularizer is much simpler than solving the L (0) regularizer. The experiments show that the L (1/2) regularizer is very useful and efficient, and can be taken as a representative of the L (p) (0 > p > 1)regularizer.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据