Second-Order Approximation of Minimum Discrimination Information in Independent Component Analysis

Independent Component Analysis (ICA) is intended to recover the mutually independent sources from their linear mixtures, and F astICA one of most successful ICA algorithms. Although it seems reasonable improve performance by introducing more nonlinear functions negentropy estimation, original fixed-point method (approximate Newton method) in degenerates under this circumstance. To alleviate problem, we propose a novel based on second-order approximation minimum discrimination information (MDI). The joint maximization our consisted minimizing single weighted least squares seeking unmixing matrix method. Experimental results validate its efficiency compared with other popular