An adaptive robust PCA neural network
Song Wang, Liang Yilong, Ma Feng · 2002
We find one way to improve the robustness of principal component analysis (PCA) based on a reconstruction error model. First, we discuss and compare the methods to analyze the robustness of the PCA algorithm. A new adaptive algorithm of robust PCA based on the structure of a single-layer neural network (NN) is developed with the 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. Simple comparison simulations are designed for verify the theoretical results.