A Greedy Layer-Wise Dictionary Training Algorithm Based on Genetic Approach

您当前的位置：

首页 >

文章列表页 >

A Greedy Layer-Wise Dictionary Training Algorithm Based on Genetic Approach

更新时间：2025-07-09

- A Greedy Layer-Wise Dictionary Training Algorithm Based on Genetic Approach
- Vol. 46, Issue 4, Pages: 18-23(2012)
- 作者机构：
  
  1. 西安交通大学电子与信息工程学院,西安,710049
  2. 西安邮电学院通信与信息工程学院,西安,710121
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP391
- Online First：10 April 2012，
  
  Published：2012
- 稿件说明：
移动端阅览
A Greedy Layer-Wise Dictionary Training Algorithm Based on Genetic Approach[J]. 2012, 46(4): 18-23.
DOI：

A Greedy Layer-Wise Dictionary Training Algorithm Based on Genetic Approach[J]. 2012, 46(4): 18-23. DOI：

摘要

针对稀疏表示残差过大的问题

提出了采用遗传算法的分层贪婪字典训练算法. 该算法首先将数据样本变成一维信号

然后将问题划分为若干个子问题

采用贪婪算法思想分层训练字典. 为了以一定概率寻找到每一层字典的最优值

使用遗传算法来训练每一层字典

最后将每层字典级联作为最终的字典. 在训练每一层字典时

先采用号码矩阵对样本的分类进行表示

然后以平均低秩逼近的残差能量作为衡量适应度的参数

以联赛选择的方式选出优胜个体

通过单点交叉和变异方法产生新的个体. 对二值序列的稀疏表示信号重建的实验结果表明

该算法在训练样本量较小的情况下

与传统的核奇异值分解算法相比

训练得到的字典在同样的稀疏度约束下重建信噪比提高了10倍以上.

Abstract

A greedy layer-wise dictionary training algorithm based on genetic approach is presented to solve the problem of too large residual in sparse representation. The algorithm firstly changes the data samples to one-dimension vectors. Then the greedy layer-wise dictionary training algorithm is employed to separate the dictionary training problem into several sub-problems. When the proposed algorithm is used to train every dictionary layer

and a genetic approach is used to find the optimal solution in every dictionary layer with a high probability. Finally

each dictionary layer is connected to get the final dictionary. When one dictionary layer is trained

matrices with numbers are used to describe the classes. Then

the average residual energy of low rank approximations is used as a measure of fitness

and Winners are selected by matching. New individuals are generated by single point crossover and mutation. Experiments about the sparse representation of the short binary sequences show that the reconstruction SNR of the proposed algorithm is 10 or more times higher than that of traditional kernel singular value decomposition algorithms under the same sparsity constraint when the number of training data samples is small.

关键词

Keywords

references

RUBINSTEIN R, BRUCKSTEIN A M, ELAD M. Dictionaries for sparse representation modeling [J]. Proceedings of the IEEE, 2010, 98(6): 1045-1057.

CHEN S, DONOHO D, SAUNDERS M. Atomic decomposition by basis pursuit [J]. SIAM Journal on Scientific Computing, 1999, 20(1): 33-61.

AHARON M, ELAD M, BRUCKTEIN A. K-SVD: an algorithm for designing overcomplete dictionaries for sparse representation [J]. IEEE Transactions on Signal Processing, 2006, 54(11): 4311-4322.

LEE H, BATTLE A, RAINA R,et al.Efficient sparse coding algorithms [J]. Advances in Neural Information Processing Systems, 2007, 19(1), 801-808.

MAIRAL J, BACH F, PONCE J, et al. Online dictionary learning for sparse coding [C]∥Proceeding of ICML'09. Montreal, Canada: Association for Computing Machinery, 2009: 689-696.

MAIRAL J, SAPIRO G, ELAD M. Multiscale sparse image representation with learned dictionaries [C]∥Proceedings of ICIP'07. San Antonio, TX, USA: Society for Industrial and Applied Mathematics Publications, 2007: 214-241.

RUBINSTEIN R, ZIBULEVSKY M, ELAD M. Double sparsity: learning sparse dictionaries for sparse signal approximation [J]. IEEE Transactions of Signal Processing, 2010, 58(3):1553-1564.

XU Zongben, SUN Jian. Image inpainting by patch propagation using patch sparsity [J]. IEEE Transactions on Image Processing, 2010, 19(5): 1153-1165.

王国栋, 阳建宏, 黎敏, 等. 基于自适应稀疏表示的宽带噪声去除算法 [J].仪器仪表学报,2011,32(8):1818-1823.

WANG Guodong, YANG Jianhong, LI Min, et al. Wideband noise removing algorithm based on adaptive sparse representation [J].Chinese Journal of Scientific Instrument, 2011, 32(8):1818-1823.

冯亮, 王平, 许廷发, 等. 运动模糊退化图像的双字典稀疏复原 [J]. 光学精密工程, 2011, 19(8):1982-1989.

FENG Lang, WANG Ping, XU Tingfa, et al. Dual dictionary sparse restoration of blurred images [J]. Optics and Precision Engineering, 2011, 19(8):1982-1989.

AHARON M, ELAD M, BRUCKTEIN A. On the uniqueness of overcomplete dictionaries, and a practical way to retrieve them [J]. Linear Algebra and Its Applications, 2006, 416(1): 48-67.

张可村,李换琴. 工程优化方法及应用 [M].西安:西安交通大学出版社,2007:238-246.

张贤达.矩阵分析与应用 [M].北京:清华大学出版社,2004:352-358.

李敏强,寇纪凇.遗传算法的基本理论与应用 [M].北京:科学出版社,2002: 120-133.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Optimized Epithermal Neutron Field Design Based on D-D Neutron Source

Optimization Design of Epithermal Neutron Field Based on D-D Neutron Source

Estimation of Failure Intensity and Maintenance Effect for Petrochemical Equipment

Component Placement Process Optimization for Dual-Gantry Placement Machine Based on Hybrid Genetic Algorithm

Pareto Optimality Based Genetic Algorithm in Web Services Composition

Related Author

JIANG Quanxu

HAN Zheng

LIN Xingyang

YANG Shanghui

DAI Yuqi

ZHANG Tianjiao

CHEN Tian

MA Baolong

Related Institution

华硼中子科技（杭州）有限公司，310000，杭州Hua Boron Neutron Technology（Hangzhou）Co., Ltd., Hangzhou 310000, China

西安交通大学湖州中子科学实验室，313000，浙江湖州Huzhou Neutron Science Laboratory, Xi'an Jiaotong University, Huzhou, Zhejiang 313000, China

西安交通大学核科学与技术学院，710049，西安School of Nuclear Science and Technology, Xi'an Jiaotong University, Xi'an 710049, China

HBNCT Co., Ltd.,Shangcheng District

School of Nuclear Science and Technology, Xi’an Jiaotong University

⁰