| ICA的梯度下降算法框架(英文) |
| |
| 引用本文： | 罗一涵,付承毓,舒勤.ICA的梯度下降算法框架(英文)[J].光电工程,2009,36(9). |
| |
| 作者姓名： | 罗一涵 付承毓 舒勤 |
| |
| 作者单位： | 1. 中国科学院光电技术研究所,成都,610209
2. 四川大学电气信息学院,成都,610065 |
| |
| 摘 要： | 为了设计更多有效的独立分量分析(ICA)算法,本文提出了ICA梯度下降算法(GDA)的一般框架,覆盖了许多目前流行的算法,如Infomax,MMI,MLE等等.该框架由一种新的基于Ⅱ类超加(减)性函数的参比函数理论导出,并采用推广的EASI形式作为更新规则来获得更好的性能.同时本丈也展示了一个基于二次熵函数的框架使用例子,并提出了其梯度的快速计算方法,最后仿真证明了它的有效性.实验结果表明,该框架非常实用,可作为开发更多有效ICA算法的有利工具.
|
| 关 键 词： | 盲信号处理 独立分量分析 梯度下降算法 串行矩阵更新 等变自适应分离 |
Framework of Gradient Descent Algorithms for ICA |
| |
| LUO Yi-han,FU Cheng-yu,SHU Qin.Framework of Gradient Descent Algorithms for ICA[J].Opto-Electronic Engineering,2009,36(9). |
| |
| Authors: | LUO Yi-han FU Cheng-yu SHU Qin |
| |
| Affiliation: | LUO Yi-han1,FU Cheng-yu1,SHU Qin2 ( 1. Institute of Optics and Electronics,Chinese Academy of Sciences,Chengdu 610209,China,2. College of Electrical Engineering and Information,Sichuan University,Chengdu 610065,China ) |
| |
| Abstract: | To design more effective algorithms for Independent Component Analysis (ICA), a general framework of Gradient Descent Algorithms (GDAs) for ICA was proposed, which covers many popular algorithms such as Infomax, Minimization of Mutual Information (MMI), Maximum Likelihood Estimation (MLE) and so on. This framework was derived from a new theory of the contrast functions for ICA based on the superadditive (or subadditive) function of class II. For better performances, the Equivariant Adaptive Separation via I... |
| |
| Keywords: | blind signal processing independent component analysis gradient descent algorithm serial matrix updating equivariant adaptive separation |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
