In this paper a hybrid approach, based on a functional net- work and a neural netw ork, for post-nonlinear independeit component analysis is presented. In order to obtain the independence among the outputs, it was used as cost function a measure based on Renvi's quadratic entrop y and Caulay-Sc h w artz inequalyit Also, the Kernel method was iLsed for nonparametric estimation of the probability dcnsit v function. The experimental results corroborated the soundness of the approach and a comparative study with a neural nctw ork sho ed its superior performance. © Springer-Verlag Berlin Heidelberg 2001.
CITATION STYLE
Romero, O. F., Berdinas, B. G., & Betanzos, A. A. (2001). A functional-neural network for post-nonlinear independent component analysis. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2084 LNCS, pp. 301–307). Springer Verlag. https://doi.org/10.1007/3-540-45720-8_34
Mendeley helps you to discover research relevant for your work.