Metode Monte Carlo za igranje iger

diplomsko delo

Mitja Rešek (Author), Damjan Strnad (Mentor), Nikola Guid (Co-mentor)

Abstract

Metode Monte Carlo so dokaj nov način uporabe umetne inteligence za igranje iger. Čeprav so se prvi koncepti uporabe te metode pojavili že zelo zgodaj, je do uporabe v igrah preteklo precej časa. V tem delu predstavimo enega izmed možnih načinov uporabe teh metod v igri dama. Metode Monte Carlo se relativno enostavno prilagajajo tej igri, saj gre za igro s popolno informacijo. Izkaže se, da je metoda tudi s pomanjkljivo hevristično funkcijo konkurenčna povprečnemu človeškemu igralcu.

Keywords

metoda Monte Carlo;igra dama;popolna informacija;drevesno preiskovanje Monte carlo;

Data

Language:	Slovenian
Year of publishing:	2013
Typology:	2.11 - Undergraduate Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	[M. Rešek]
UDC:	004.8:794.98(043.2)
COBISS:	17493526
Views:	1584
Downloads:	111
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	MONTE-CARLO METHODS FOR GAME PLAYING
Secondary abstract:	Monte-Carlo methods are a new way of using artificial intelligence for game playing. First concepts of using this method appeared quite some time ago, but first use in gaming came a lot later. In the work we present one of possible uses of that algorithm in the game of checkers. Monte-Carlo methods are relatively simple to adapt for that game, because it’s a game with perfect information. It shows that even with simple heuristic function the method is competitive to average human players.
Secondary keywords:	Monte carlo methods;checkers;games with perfect information;Monte carlo tree search;
URN:	URN:SI:UM:
Type (COBISS):	Bachelor thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko
Pages:	VI, 26 f.
ID:	8727855