【机器学习】Nonlinear Independent Component Analysis - Aapo Hyvärinen
Linearindependentcomponentanalysis(ICA)xi(k)=∑j=1naijsj(k)foralli=1…n,k=1…K()x_i(k)=\sum_{j=1}^{n}a_{ij}s_j(k)\quad\text{forall}i=1\ldotsn,k=1\ldotsK\tag{}xi(k)=j=1∑naijsj(k)foralli=1…n,k=1…K()xi(k)x_