. "285;292" . "Koldovsk\u00FD, Zbyn\u011Bk" . "0302-9743" . "P(1M0572), Z(AV0Z10750506)" . "Zlep\u0161en\u00ED rychlosti a p\u0159esnosti metod anal\u00FDzy nez\u00E1visl\u00FDch komponent pomoc\u00ED racion\u00E1ln\u00EDch neline\u00E1rn\u00EDch funkc\u00ED"@cs . . "Speed and accuracy enhancement of linear ICA techniques using rational nonlinear functions"@en . . "3"^^ . "[C2C437621E0B]" . "Oja, E." . . "4666" . "Mnoho algoritm\u016F pro anal\u00FDzu nez\u00E1visl\u00FDch komponent kter\u00E9 jsou zalo\u017Een\u00E9 na minimalizaci neline\u00E1rnikontrastn\u00ED funkce pou\u017E\u00EDvaj\u00ED hyperbolickou tangentu (tanh) jako svoj\u00ED vestav\u011Bnou nelinearitu. V \u010Dl\u00E1nku je navr\u017Eeno nahrazen\u00ED tanh dv\u011Bmi racion\u00E1ln\u00EDmi funkcemi a dal\u0161\u00ED racion\u00E1ln\u00ED funkce jsou navr\u017Eeny pro separaci supergaussovskych zdroj\u016F, tj. zdroj\u016F jejich\u017E pdf ma t\u011B\u017Ek\u00E9 chvosty. V\u00FDhodou racion\u00E1ln\u00EDch funkc\u00ED je sni\u017Een\u00ED v\u00FDpo\u010Detn\u00ED slo\u017Eitosti a p\u0159\u00EDpadn\u011B t\u00E9\u017E zlep\u0161en\u00ED kvality separace, pokud jsou nelinearity vybr\u00E1ny adaptivn\u011B ke zpracov\u00E1van\u00FDm dat\u016Fm."@cs . "2"^^ . . "DE - Spolkov\u00E1 republika N\u011Bmecko" . . . "Speed and accuracy enhancement of linear ICA techniques using rational nonlinear functions" . . "451587" . "Many linear ICA techniques are based on minimizing a nonlinear contrast function and many of them use a hyperbolic tangent (tanh) as their built-in nonlinearity. In this paper we propose two rational functions to replace the tanh and other popular functions that are tailored for separating supergaussian (long-tailed) sources. The advantage of the rational function is two-fold. First, the rational function requires a significantly lower computational complexity than tanh, e.g. nine times lower. As a result, algorithms using the rational functions are typically twice faster than algorithms with tanh. Second, it can be shown that a suitable selection of the rational function allows to achieve a better performance of the separation in certain scenarios. This improvement might be systematic, if the rational nonlinearities are selected adaptively to data." . "blind source separation; independent component analysis"@en . . "Tichavsk\u00FD, Petr" . . . . "Speed and accuracy enhancement of linear ICA techniques using rational nonlinear functions" . "Speed and accuracy enhancement of linear ICA techniques using rational nonlinear functions"@en . "Many linear ICA techniques are based on minimizing a nonlinear contrast function and many of them use a hyperbolic tangent (tanh) as their built-in nonlinearity. In this paper we propose two rational functions to replace the tanh and other popular functions that are tailored for separating supergaussian (long-tailed) sources. The advantage of the rational function is two-fold. First, the rational function requires a significantly lower computational complexity than tanh, e.g. nine times lower. As a result, algorithms using the rational functions are typically twice faster than algorithms with tanh. Second, it can be shown that a suitable selection of the rational function allows to achieve a better performance of the separation in certain scenarios. This improvement might be systematic, if the rational nonlinearities are selected adaptively to data."@en . "Lecture Notes in Computer Science" . . . "RIV/67985556:_____/07:00085697!RIV08-AV0-67985556" . "2007" . . . "8"^^ . "RIV/67985556:_____/07:00085697" . . "Zlep\u0161en\u00ED rychlosti a p\u0159esnosti metod anal\u00FDzy nez\u00E1visl\u00FDch komponent pomoc\u00ED racion\u00E1ln\u00EDch neline\u00E1rn\u00EDch funkc\u00ED"@cs .