Thesis Details
Evolutionary Design Using Grammatical Evolution
p, li { white-space: pre-wrap; } Natural evolution serves as a source of inspiration for this thesis. The basic algorithm utilizes generational power of grammars in combination with evolutionary approach. The search for behavior strategies in different environments draws from evolutionary methods. Behavior trees are the model generally used to control decision making of some artificial intelligence. This thesis seeks for behavior trees which would control individuals solving the following two problems: an adjusted version of knight's tour problem and playing the game Liar's dice. When searching for a strategy of a player in a game, a competitive coevolution was implemented to mitigate the difficulty of designing a good fitness function.
p, li { white-space: pre-wrap; } grammatical evolution, genetic algorithm, competitive coevolution, behavior trees, AI control, knight's tour, Liar's dice
Burgetová Ivana, Ing., Ph.D. (DIFS FIT BUT), člen
Horák Aleš, doc. RNDr., Ph.D. (FI MUNI), člen
Martínek Tomáš, doc. Ing., Ph.D. (DCSY FIT BUT), člen
Rogalewicz Adam, doc. Mgr., Ph.D. (DITS FIT BUT), člen
Zbořil František, doc. Ing., Ph.D. (DITS FIT BUT), člen
@mastersthesis{FITMT19174, author = "Tom\'{a}\v{s} Rep\'{i}k", type = "Master's thesis", title = "Evolutionary Design Using Grammatical Evolution", school = "Brno University of Technology, Faculty of Information Technology", year = 2017, location = "Brno, CZ", language = "english", url = "https://www.fit.vut.cz/study/thesis/19174/" }