摘要
研究了改善主成分分析（ＰＣＡ）算法鲁棒性的一种实现途径．通过对误差函数的建模分析，得到一种改进的目标函数．提出一种新的在线自适应式的鲁棒ＰＣＡ运算规则．该方法基于单层线性神经网络（ＮＮ）结构，但是权值的训练算法是非线性的．从而在迭代训练中对“劣点”样本加以适当处理来排除对运算精度和收敛性的影响．
One way to improve the robustness of principal component analysis (PCA) is studied in the paper. A new adaptive algorithm of robust PCA based on the structure of single layer neural network (NN) is developed with modification of the cost function which can be acquired through modeling of the error function. The new nonlinear robust PCA algorithm can reduce the effects of outliers on the accuracy and convergence of the PCA algorithm through proper processing of them.
出处
《自动化学报》
EI
CSCD
北大核心
1999年第4期528-531,共4页
Acta Automatica Sinica
基金
国家自然科学基金
关键词
主成分分析
鲁棒性
误差模型
协方差分析
Principal component analysis (PCA), adaptive robust PCA, outliers, neural network (NN), maximum likelihood estimate.
