. "2010-12-31+01:00"^^ . "2015-03-20+01:00"^^ . . . . " evolutionary algorithms" . . "Metody strojov\u00E9ho u\u010Den\u00ED pro konstrukci \u0159e\u0161en\u00ED v evolu\u010Dn\u00EDch algoritmech" . "     The project was aimed at using machine learning methods for solution construction in evolutionary algorithms (EAs) with real-valued representation. It studied the use of probabilistic models in the role of reproduction operators in the framework of estimation-of-distribution algorithms (EDAs) and the the use of classification models in the framework of the so-called learn"@en . . "2010-04-16+02:00"^^ . . . "0"^^ . "0"^^ . "     Projekt byl zam\u011B\u0159en na vyu\u017Eit\u00ED metod strojov\u00E9ho u\u010Den\u00ED ke konstrukci \u0159e\u0161en\u00ED v evolu\u010Dn\u00EDch algoritmech s re\u00E1lnou reprezentac\u00ED. Studoval p\u0159edev\u0161\u00EDm vyu\u017Eit\u00ED pravd\u011Bpodobnostn\u00EDch model\u016F v roli reproduk\u010Dn\u00EDch oper\u00E1tor\u016F v r\u00E1mci tzv. estimation-of-distribution algorithms (EDA) a vyu\u017Eit\u00ED klasifika\u010Dn\u00EDch model\u016F v r\u00E1mci tzv. learnable evolution model (LEM). V\u0161echny pr\u00E1ce na projektu byly"@cs . . "1"^^ . . . "Machine learning methods for solution construction in evolutionary algorithms"@en . "V klasick\u00FDch evolu\u010Dn\u00EDch algoritmech (EA) zalo\u017Een\u00FDch na k\u0159\u00ED\u017Een\u00ED a mutaci je okol\u00ED, z n\u011Bho\u017E se generuj\u00ED nov\u00E1 \u0159e\u0161en\u00ED, statick\u00E9 a je d\u00E1no implicitn\u011B, a to pr\u00E1v\u011B pou\u017Eit\u00FDmi oper\u00E1tory. V r\u00E1mci EA se ji\u017E dlouho objevuj\u00ED snahy pou\u017E\u00EDvat okol\u00ED, kter\u00E9 by se bylo schopn\u00E9 adaptovat na jednotliv\u00E9 oblasti prohled\u00E1van\u00E9ho prostoru a um\u011Blo by modelovat i interakce mezi jednotliv\u00FDmi rysy kvalitn\u00EDch \u0159e\u0161en\u00ED. Tyto snahy vy\u00FAstily ve vznik tzv. estimation of distribution algorithms (EDA), kter\u00E9 pro definici okol\u00ED pou\u017E\u00EDvaj\u00ED explicitn\u011B vytvo\u0159en\u00E9 pravd\u011Bpodobnostn\u00ED modely, z nich\u017E se n\u00E1sledn\u011B vzorkuj\u00ED nov\u00ED kandid\u00E1ti na \u0159e\u0161en\u00ED optimaliza\u010Dn\u00ED \u00FAlohy. Jin\u00FD zp\u016Fsob vytv\u00E1\u0159en\u00ED pravd\u011Bpodobnostn\u00EDch model\u016F obsahuj\u00ED algoritmy typu learnable evolution model (LEM), kter\u00E9 nejprve vytv\u00E1\u0159ej\u00ED klasifik\u00E1tor rozli\u0161uj\u00EDc\u00ED kvalitn\u00ED a m\u00E9n\u011B kvalitn\u00ED jedince a n\u00E1sledn\u011B popis t\u011Bch kvalitn\u00EDch p\u0159evedou do formy pravd\u011Bpodobnostn\u00EDho modelu. Tento projekt se zam\u011B\u0159\u00ED na prohlouben\u00ED poznatk\u016F o t\u011Bchto dvou typech algoritm\u016F, a to sou\u010Dasn\u011B s p\u0159ihl\u00E9dnut\u00EDm k" . "12"^^ . . "12"^^ . . "optimization; evolutionary algorithms; machine learning; population diversity"@en . " machine learning" . . . "GP102/08/P094" . "http://www.isvav.cz/projectDetail.do?rowId=GP102/08/P094"^^ . "2008-01-01+01:00"^^ . . . . "optimization" . . "The neighborhood used in conventional evolutionary algorithms (EA) to create new candidate problem solutions is static and is given implicitly by the crossover and mutation operators used in the algorithm. In the field of EA, many researchers have tried to use neighborhoods able to adapt to certain areas of the search space and to cover some interactions among the features of the promising solutions. The result is the establishment of the so-called estimation of distribution algorithms (EDA) which use probabilistic models to describe the neighborhood. The model is explicitly built on the basis of promising individuals and is used to sample new candidate solutions. Somewhat different approach for the creation of probabilistic model is presented in the so-called learnable evolution model (LEM): first, a classifier distinguishing between promising and less promising solutions is created and then the description of promising solutions is turned into a probabilistic model. This project shall deepen our"@en .