"1"^^ . "Projekt p\u0159isp\u011Bl k matematick\u00E9 teorii slo\u017Eitosti neuronov\u00FDch s\u00EDt\u00ED. Byly odvozeny odhady modelov\u00E9 slo\u017Eitosti v z\u00E1vislosti na dimenzi vstupn\u00EDch dat, typech a parametrech v\u00FDpo\u010Detn\u00EDch jednotek, vlastnostech aproximovan\u00FDch funkc\u00ED a tr\u00E9novac\u00EDch dat. Byly navr\u017Eeny evolu\u010Dn\u00ED algoritmy metau\u010Den\u00ED, realizov\u00E1na jejich implementace jako softwarov\u00E9ho n\u00E1stroje a ov\u011B\u0159ena funk\u010Dnost a praktick\u00E9 vlastnosti."@cs . . "ME10023" . "0"^^ . . . "0"^^ . . "C\u00EDlem projektu je odvozen\u00ED teoretick\u00FDch v\u00FDsledk\u016F a n\u00E1vrh a softwarov\u00E1 realizace u\u010D\u00EDc\u00EDch algoritm\u016F pro modely neurov\u00FDpo\u010Dt\u016F zalo\u017Een\u00FDch na aproximaci funkc\u00ED v\u00EDce prom\u011Bnn\u00FDch." . "Project contributed to mathematical theory of complexity of neural networks. Estimates of model complexity in dependence on dimension of input data, types and parameters of computational units, properties of approximated functions and training data were derived. Evolutionary algorithms for metalearning were designed, implemented as a software tool, and functionality and properties were verified."@en . "http://www.isvav.cz/projectDetail.do?rowId=ME10023"^^ . . . "6"^^ . "6"^^ . . "neural networks" . . " approximation of functions" . . "2013-06-28+02:00"^^ . "Research of approximation and learning from data of multivariable functions by neural networks and design of learning algorithms"@en . "Aproximation and learning of multivariable functions by neural networks"@en . "2012-02-16+01:00"^^ . "2012-12-31+01:00"^^ . . . . . . . . . "Aproximace a u\u010Den\u00ED funkc\u00ED mnoha prom\u011Bnn\u00FDch pomoc\u00ED neuronov\u00FDch s\u00EDt\u00ED" . "neural networks; approximation of functions; learning algorithms"@en . . "2010-05-01+02:00"^^ . .