| 一种用于独立分量分析的准最大期望算法研究 |
| |
| 引用本文： | 陈菲,王刚.一种用于独立分量分析的准最大期望算法研究[J].四川大学学报(自然科学版),2008,45(6):1347-1351. |
| |
| 作者姓名： | 陈菲 王刚 |
| |
| 作者单位： | 1. 西南科技大学计算机科学与技术学院,绵阳,621010
2. 电子科技大学生命科学与技术学院,成都,610054 |
| |
| 基金项目： | 国家自然科学基金(60274022 F 030112) |
| |
| 摘 要： | 对于给定的独立分量分析（ICA）的对比函数，提出了一种准最大期望学习算法及其迭代方法，这个算法研究了在批处理方法下寻找最优解，并用类似于证明EM算法收敛的辅助函数证明了该算法的收敛性.计算机仿真表明，该算法比Past-ICA算法鲁棒性好.
|
| 关 键 词： | 独立分量分析 对比函数 准最大期望 峭度 |
A quasi expectation-maximization algorithm for independent component analysis |
| |
| CHEN Fei and WANG Gang.A quasi expectation-maximization algorithm for independent component analysis[J].Journal of Sichuan University (Natural Science Edition),2008,45(6):1347-1351. |
| |
| Authors: | CHEN Fei and WANG Gang |
| |
| Institution: | Computer Science and Technology,Southwest University Science and Technology;School of life Science and Technology, University of Electronic Science and Technology |
| |
| Abstract: | Given the contrast function of independent component analysis(ICA),the authors focus our research on how to optimize it with batch method.In this paper,the authors show how a quasi Expectation-Maximization(QEM) learning rule is transformed into iterations.The convergence of the algorithm can be proven using auxiliary function analogous to that used for proving convergence of the EM algorithm.The algorithm can also be interpreted as diagonally rescaled gradient descent.The authors demonstrate the operation o... |
| |
| Keywords: | independent component analysis contract function quasi expectation-maximization kurtosis |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《四川大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《四川大学学报(自然科学版)》下载免费的PDF全文 |
