INDEPENDENT COMPONENT ANALYSIS

• ICA is about factoring probability distributions, and doing blind source separation

  
  

• ICA is related to entropy and information maximisation, maximum likelihood density estimation (MLE), EM (expectation maximisation, which is MLE with hidden variables) and projection pursuit.

  
  

• ICA is a way of finding special linear (non-orthogonal) co-ordinate systems in multivariate data, using higher-order statistics in various ways.

  
  

• Similar to PCA, ICA determines a linear co-ordinate system for the data, transforming it by matrix W into a new basis set.

  
  

• The transformation is free to be non-orthogonal and the axes is sensitive to statistics of all orders as against PCA which is sensitive to the second order statistics.