Correlated topographic analysis estimates non-Gaussian components and their ordering from data. The ordering is determined based on both linear and energy correlations.

MATLAB implementation of CTA:  CTA.zip

References

• Hiroaki Sasaki, Michael U. Gutmann, Hayaru Shouno and Aapo Hyvärinen, “Correlated Topographic Analysis: Estimating an Ordering of Correlated Components”, Machine Learning, vol.92, no.2-3, pp.285-317, 2013.

• Hiroaki Sasaki, Michael U. Gutmann, Hayaru Shouno and Aapo Hyvärinen, “Topographic Analysis of Correlated Components”, the4th Asian Conference on Machine learning (ACML), JMLR: Workshop and Conference Proceedings, vol.25, pp.365-378, 2012.