. . . . "genetick\u00FDch algoritm\u016F"@cs . . . "Genetick\u00FD algoritmus"@cs . "Genetick\u00FD algoritmus (GA) je heuristick\u00FD postup, kter\u00FD se sna\u017E\u00ED aplikac\u00ED princip\u016F evolu\u010Dn\u00ED biologie nal\u00E9zt \u0159e\u0161en\u00ED slo\u017Eit\u00FDch probl\u00E9m\u016F, pro kter\u00E9 neexistuje pou\u017Eiteln\u00FD exaktn\u00ED algoritmus. Genetick\u00E9 algoritmy, resp. v\u0161echny postupy pat\u0159\u00EDc\u00ED mezi tzv. evolu\u010Dn\u00ED algoritmy, pou\u017E\u00EDvaj\u00ED techniky napodobuj\u00EDc\u00ED evolu\u010Dn\u00ED procesy zn\u00E1m\u00E9 z biologie \u2013 d\u011Bdi\u010Dnost, mutace, p\u0159irozen\u00FD v\u00FDb\u011Br a k\u0159\u00ED\u017Een\u00ED \u2013 pro \u201E\u0161lecht\u011Bn\u00ED\u201C \u0159e\u0161en\u00ED zadan\u00E9 \u00FAlohy.Princip pr\u00E1ce genetick\u00E9ho algoritmu je postupn\u00E1 tvorba generac\u00ED r\u016Fzn\u00FDch \u0159e\u0161en\u00ED dan\u00E9ho probl\u00E9mu. P\u0159i \u0159e\u0161en\u00ED se uchov\u00E1v\u00E1 tzv. populace, jej\u00ED\u017E ka\u017Ed\u00FD jedinec p\u0159edstavuje jedno \u0159e\u0161en\u00ED dan\u00E9ho probl\u00E9mu. Jak populace prob\u00EDh\u00E1 evoluc\u00ED, \u0159e\u0161en\u00ED se zlep\u0161uj\u00ED. Tradi\u010Dn\u011B je \u0159e\u0161en\u00ED reprezentov\u00E1no bin\u00E1rn\u00EDmi \u010D\u00EDsly, \u0159et\u011Bzci nul a jedni\u010Dek, nicm\u00E9n\u011B pou\u017E\u00EDvaj\u00ED se i jin\u00E9 reprezentace (strom, pole, matice, \u2026). Typicky je na za\u010D\u00E1tku simulace (v prvn\u00ED generaci) populace slo\u017Eena z naprosto n\u00E1hodn\u00FDch \u010Dlen\u016F. V p\u0159echodu do nov\u00E9 generace je pro ka\u017Ed\u00E9ho jedince spo\u010Dtena tzv. fitness funkce, kter\u00E1 vyjad\u0159uje kvalitu \u0159e\u0161en\u00ED reprezentovan\u00E9ho t\u00EDmto jedincem. Podle t\u00E9to kvality jsou stochasticky vybr\u00E1ni jedinci, kte\u0159\u00ED jsou modifikov\u00E1ni (pomoc\u00ED mutac\u00ED a k\u0159\u00ED\u017Een\u00ED), \u010D\u00EDm\u017E vznikne nov\u00E1 populace. Tento postup se iterativn\u011B opakuje, \u010D\u00EDm\u017E se kvalita \u0159e\u0161en\u00ED v populaci postupn\u011B vylep\u0161uje. Algoritmus se obvykle zastav\u00ED p\u0159i dosa\u017Een\u00ED posta\u010Duj\u00EDc\u00ED kvality \u0159e\u0161en\u00ED, p\u0159\u00EDpadn\u011B po p\u0159edem dan\u00E9 dob\u011B."@cs . "Genetick\u00FD algoritmus"@cs . . "genetick\u00FDmi algoritmy"@cs . . "27"^^ . . . . "10965"^^ . . . . "Genetick\u00FD algoritmus (GA) je heuristick\u00FD postup, kter\u00FD se sna\u017E\u00ED aplikac\u00ED princip\u016F evolu\u010Dn\u00ED biologie nal\u00E9zt \u0159e\u0161en\u00ED slo\u017Eit\u00FDch probl\u00E9m\u016F, pro kter\u00E9 neexistuje pou\u017Eiteln\u00FD exaktn\u00ED algoritmus. Genetick\u00E9 algoritmy, resp. v\u0161echny postupy pat\u0159\u00EDc\u00ED mezi tzv."@cs . . . . . . . . . . "15391742"^^ . . . . . . . . . . "15337"^^ . "genetick\u00E9 algoritmy"@cs . . . . . .