Independent Slow Feature Analysis and Nonlinear Blind Source Separation

We present independent slow feature analysis as a new method for nonlinear blind source separation. It circumvents the indeterminacy of nonlinear independent component analysis by constraining the independent components to be as slowly varying as possible. This is achieved by incorporating the principle of slow feature analysis into a nonlinear independent component analysis algorithm based on second-order statistics. The performance of the algorithm is demonstrated on nonlinearly mixed speech data.

• Type: Conference
• Year: 2004
• Where: ICA
• Authors: Tobias Blaschke, Laurenz Wiskott
• Keywords: ICA 2004, Independent Component Analysis, Nonlinear Independent Component, Slow Feature Analysis