Mutual information minimization: application to Blind Source Separation

In this paper, the problem of Blind Source Separation (BSS) through mutual information minimization is addressed. For mutual information minimization, multi-variate score functions are first introduced, which can be served to construct a non-parametric “gradient” for mutual information. Then, two general gradient based approaches for minimizing mutual information in a parametric model are presented. Although, in this paper, these approaches are only used in BSS, they are quiet general, and can be applied in other mutual information optimization problems. Index Terms Mutual Information, Information theoretic learning, Gradient of mutual information, Independent Component Analysis (ICA), Blind Source Separation (BSS).