izpis_h1_title_alt

Preiskovanje pri igrah z nepopolno informacijo na primeru tršeta : magistrsko delo
ID Kafol, Žan (Author), ID Robnik Šikonja, Marko (Mentor) More about this mentor... This link opens in a new window

URLURL - Presentation file, Visit http://eprints.fri.uni-lj.si/2819/ This link opens in a new window

Abstract
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.

Language:Slovenian
Keywords:preiskovanje dreves Monte Carlo, nepopolna informacija, umetna inteligenca, tršet, ekspektiminimaks, preiskovanje, računalništvo, računalništvo in informatika, magisteriji
Work type:Master's thesis/paper
Typology:2.09 - Master's Thesis
Organization:FRI - Faculty of Computer and Information Science
Publisher:[Ž. Kafol]
Year:2014
Number of pages:40 str.
PID:20.500.12556/RUL-69054 This link opens in a new window
UDC:004.83(043.2)
COBISS.SI-ID:1536017603 This link opens in a new window
Publication date in RUL:10.07.2015
Views:1148
Downloads:256
Metadata:XML RDF-CHPDL DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Search in incomplete information games: a case of tressete
Abstract:
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.

Keywords:Monte Carlo tree search, incomplete information, artificial inteligence, Tressette, expectiminimax, search algorithm, game playing, computer science, computer and information science, master's degree

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back