| 含隐变量非高斯无环因果模型的估计算法 |
| |
| 引用本文： | 姜枫,朱辉生,汪卫.含隐变量非高斯无环因果模型的估计算法[J].计算机工程,2010,36(9):178-180. |
| |
| 作者姓名： | 姜枫 朱辉生 汪卫 |
| |
| 作者单位： | 1. 南京理工大学泰州科技学院计算机科学与技术系,南京,225300
2. 复旦大学计算机科学技术学院,上海,200433 |
| |
| 基金项目： | 国家自然科学基金资助项目(60303008);;国家“973”计划基金资助项目(2005CB321905) |
| |
| 摘 要： | 针对观测变量中含隐变量的非高斯线性无环因果模型的估计问题，提出一种新的算法。通过在超完备基独立成分分析算法中引入满足Oracle性质的惩罚因子，使混合矩阵的估计值具有稀疏连接权值，由此推导出模型估计算法。实验结果表明，该算法能够改进因果模型估计的精确程度，提高算法效率。
|
| 关 键 词： | 线性因果模型 超完备基 独立成分分析 稀疏连接 |
| 修稿时间： | |
Estimation Algorithm for Non-Gaussian Acyclic Causal Model with Latent Variables |
| |
| JIANG Feng,ZHU Hui-sheng,WANG Wei.Estimation Algorithm for Non-Gaussian Acyclic Causal Model with Latent Variables[J].Computer Engineering,2010,36(9):178-180. |
| |
| Authors: | JIANG Feng ZHU Hui-sheng WANG Wei |
| |
| Affiliation: | (1. Dept. of Computer Science and Technology, Taizhou Institute of Science, Nanjing University of Science and Technology, Nanjin 225300; 2. School of Computer Science, Fudan University, Shanghai 200433) |
| |
| Abstract: | In order to estimate the non-Gaussian linear acyclic causal models with latent variables, a new algorithm is proposed. The algorithm is derived by integrating penalty functions which satisfy the Oracle property into overcomplete Independent Component Analysis(ICA), making the entries of mixing matrix sparse. Experimental results show that the algorithm improves accuracy and efficiency of the casual model estimation. |
| |
| Keywords: | linear causal model overcomplete basis Independent Component Analysis(ICA) sparse connection |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《计算机工程》浏览原始摘要信息 |
|
点击此处可从《计算机工程》下载全文 |
