An ensemble learning approach to independent component analysis
Rizwan Choudrey, W.D. Penny, Stephen John Roberts · 2002
Independent Component Analysis (ICA) is an important tool for extracting structure from data. ICA is traditionally performed under a maximum likelihood scheme in a latent variable model and in the absence of noise. Although extensively utilised maximum likelihood estimation has well known drawbacks such as overfitting and sensitivity to local-maxima. We propose a Bayesian learning scheme, Variational Bayes or Ensemble Learning, for both latent variables and parameters in the model.