20.500.12556/RUL-29965 Preiskovanje pri igrah z nepopolno informacijo na primeru tršeta magistrsko delo Search in incomplete information games: a case of tressete V magistrskem delu preizkušamo različne pristope za reševanje problema preiskovanja z nepopolno informacijo. Za primer smo izbrali igro s kartami tršet, kjer nepopolno informacijo predstavljajo karte v kupčku, to pa pomeni, da možne poteze igralcem niso vidne in na igro vpliva verjetnost. Glavni poudarek je na metodi preiskovanja dreves Monte Carlo (MCTS), ki temelji na naključnih simulacijah in preišče le del prostora. MCTS se je na tej domeni izkazal za uspešno metodo. Razvili smo prototip avtomatskega agenta za igranje igre, ga postopoma izboljševali s spreminjanjem parametrov ter vpeljevanjem hevristik ter merili njegovo uspešnost. V preiskovanje smo vključili tudi znanje, pridobljeno iz baze človeških iger ter testirali vpliv parametrov na uspešnost. Uspešnost smo ovrednotili na podlagi iger proti človeškim igralcem ter z medsebojnim igranjem različnih pristopov. We explore different approaches for solving problems with incomplete information. As an example a card game Tressette is chosen where the incomplete information is presented as cards still in the deck. This means that players cannot make deterministic strategies on possible outcomes or predict the moves of an opponent, because such moves are not guaranteed, but are possible with certain probability. The main emphasis is on the Monte Carlo tree search method (MCTS), which uses random sampling and simulates only a part of the search space. MCTS has proven to be a successful method in this domain. A prototype of an intelligent agent was developed for playing the game. The agent was gradually improved by tuning MCTS method parameters and by introducing new heuristics into the search. We used knowledge extracted from the database of human-played games in the agent to improve its efficiency. The agent was tested by different approaches playing against each other and against human players. Preiskovanje dreves Monte Carlo nepopolna informacija umetna inteligenca tršet ekspektiminimaks preiskovanje računalništvo računalništvo in informatika magisteriji Monte Carlo Tree Search Incomplete information Artificial Intelligence Tressette Expectiminimax search algorithm Tressette game playing computer science computer and information science master's degree true false false Ž. Kafol Slovenski jezik Angleški jezik Magistrsko delo/naloga 2014-10-03 09:30:01 2014-10-03 09:30:02 2024-08-13 13:49:14 0000-00-00 00:00:00 2014 0 0 40 str. 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 13995 1536017603 Kafol_Zan_-_Preiskovanje_pri_igrah_z_nepopolno_informacijo_na_primeru_trseta.pdf Kafol_Zan_-_Preiskovanje_pri_igrah_z_nepopolno_informacijo_na_primeru_trseta.pdf 1 5E04809389F208F61E8C35B94F5A3026 de28772b8a16b1bc8e7a78b0bdf1746c0f64385a54ac79c35103287275b831e5 30116224-a1ad-11eb-a523-00155dcfd717 20.500.12556/rul/dc25c81a-6dd0-42e3-9e30-843e4196ce18 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=29960 Fakulteta za računalništvo in informatiko 0 0 0