| E. Oja. PCA, ICA and nonlinear Hebbian learning. In Proceedings of the ICANN'95, pages 80-94, 1995. |
....higher order statistics. There exists several neural approaches to the non linear PCA, like the well known Herault Jutten algorithm. An exhaustive listing of the relevant algorithms is given in [3, 4] In this work, we chose the Non linear PCA subspace rule, introduced by Oja in 1991 (see also [8, 7]) as a non linear extension of Oja s PCA subspace rule. We apply this rule by first pre whiten the input data. Pre whitening removes the effects of secondorder statistics from the input data. We use the standard PCA as pre whitening transformation, which we have described in the former section ....
E. Oja, PCA, ICA and Nonlinear Hebbian Learning, in Proc. of the ICANN'95, Paris, France, Oct. 9-13, 1995
No context found.
E. Oja. PCA, ICA and nonlinear Hebbian learning. In Proceedings of the ICANN'95, pages 80-94, 1995.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC