|
|
Heuristics for the Scotland Yard Board Game
Cejpek, Michal ; Zbořil, František (referee) ; Zbořil, František (advisor)
This thesis explores the possibility of using deep and reinforcement learning algorithms to solve problems with incomplete information. The main algorithm under investigation is PPO – Proximal Policy Optimization. In order to test the suitability of the PPO algorithm, a simplified implementation of the Scotland Yard game was created as well as an environment for training and testing the algorithms. From performed experiments, it emerged that the PPO algorithm is very suitable for solving problems with incomplete information. The agents very quickly gained a sense of the game’s goals and built appropriate strategies to meet those goals through training.
|
|
| |
|
|
Strategic Game with Uncertainity
Gerža, Martin ; Zbořil, František (referee) ; Zbořil, František (advisor)
This thesis focuses on the implementation of a system for playing the board game Scotland Yard autonomously and also focuses on a comparison of this system with similar ones. I focused on obtaining enough information about the possible methods that should be suitable for such a system and decided to implement this system using the Monte Carlo Tree Search method. The result implementation of the system was tested against similar systems, achieving an excellent result against another system that used an equivalent method. There was achieved a balanced result against a system that used the Alpha-Beta method. The main result of this work is a working version of an autonomous system for playing the game Scotland Yard on a reduced field. It also provides the possibility of using two similar systems within a single program in order to compare their implementations.
|
|
|
Strategic Game with Uncertainity
Sova, Michal ; Zbořil, František (referee) ; Zbořil, František (advisor)
The thesis focuses on creating an autonomous system for the game Scotland Yard by using machine learning method. The problem is solved by algorithm Monte Carlo tree search. Algorithm Monte Carlo tree search was tested against algorithm Alpha-beta. These results showed that Monte Carlo tree search algorithm is operational but win rate of this algorithm is lower than win rate of algorithm Alpha-beta. The resulting system is functional, autonomous and capable of playing the game Scotland Yard on simplified game area. There was an attempt to expand simplified version of the game Scotland Yard. In expanded version algorithm Alpha-beta was not successful because of insufficient computational resources. Algorithm Monte Carlo tree search, on the other hand, was more successful in expanded version.
|
|
|
Strategic Game with Uncertainity
Tulušák, Adrián ; Šimek, Václav (referee) ; Zbořil, František (advisor)
The thesis focuses on creating an autonomous functional system for the game Scotland Yard by using artificial intelligence methods for game theory and machine learning. The problem is solved by algorithm of game theory - Alpha Beta. There was an attempt to use machine learning, but it proved to be unsuccessful due to the large number of states for expansion and insufficient computational recourses. The solution using Alpha Beta algorithm was tested on human players and it proved the ability of artificial intelligence to fully compete against real players. The resulting system is functional, autonomous and capable of playing the game Scotland Yard on simplified game area. Based on these experiments, the thesis also introduces some improvements that could utilize machine learning and extend the existing solution.
|
|
|
Deep Learning Methods for Machine Playing the Scotland Yard Board Game
Hrkľová, Zuzana ; Janoušek, Vladimír (referee) ; Zbořil, František (advisor)
Táto práca sa zaoberá metódami hlbokého učenia, ktoré sú aplikovateľné na stolné hry s neurčitosťou. V rámci práce boli naštudované princípy učenia s posilňovaním, s hlavným zameraním na Q-learning algoritmy, spomedzi ktorých bol vybraný Deep Q-Network algoritmus. Ten bol následne implementovaný na zjednodušených pravidlách stolnej hry Scotland Yard. Konečná implementácia bola porovnaná s metódami Alpha-Beta a Monte Carlo Tree Search. S výsledkov vyplinulo, že schovávaný hráč riadený DQN algoritmom predstavoval pre ostatné metódy najťažšieho protihráča, narozdiel od hľadajúcich hráčov, ktorým sa nepodarilo zlepšiť existujúce riešenia. Napriek tomu, že implementovaná metóda nedosiahla lepšie výsledky oproti doposiaľ existujúcim metódam, ukázalo sa, že potrebuje najmenej výpočetných zdrojov a času na vykonanie daného ťahu. To ju robí najperspektívnejšou zo spomínaných metód na budúcu možnú implementáciu originálnej verzie danej hry.
|
|
| |
|
|
Strategic Game with Uncertainity
Gerža, Martin ; Zbořil, František (referee) ; Zbořil, František (advisor)
This thesis focuses on the implementation of a system for playing the board game Scotland Yard autonomously and also focuses on a comparison of this system with similar ones. I focused on obtaining enough information about the possible methods that should be suitable for such a system and decided to implement this system using the Monte Carlo Tree Search method. The result implementation of the system was tested against similar systems, achieving an excellent result against another system that used an equivalent method. There was achieved a balanced result against a system that used the Alpha-Beta method. The main result of this work is a working version of an autonomous system for playing the game Scotland Yard on a reduced field. It also provides the possibility of using two similar systems within a single program in order to compare their implementations.
|
|
|
Strategic Game with Uncertainity
Sova, Michal ; Zbořil, František (referee) ; Zbořil, František (advisor)
The thesis focuses on creating an autonomous system for the game Scotland Yard by using machine learning method. The problem is solved by algorithm Monte Carlo tree search. Algorithm Monte Carlo tree search was tested against algorithm Alpha-beta. These results showed that Monte Carlo tree search algorithm is operational but win rate of this algorithm is lower than win rate of algorithm Alpha-beta. The resulting system is functional, autonomous and capable of playing the game Scotland Yard on simplified game area. There was an attempt to expand simplified version of the game Scotland Yard. In expanded version algorithm Alpha-beta was not successful because of insufficient computational resources. Algorithm Monte Carlo tree search, on the other hand, was more successful in expanded version.
|
|
|
Strategic Game with Uncertainity
Tulušák, Adrián ; Šimek, Václav (referee) ; Zbořil, František (advisor)
The thesis focuses on creating an autonomous functional system for the game Scotland Yard by using artificial intelligence methods for game theory and machine learning. The problem is solved by algorithm of game theory - Alpha Beta. There was an attempt to use machine learning, but it proved to be unsuccessful due to the large number of states for expansion and insufficient computational recourses. The solution using Alpha Beta algorithm was tested on human players and it proved the ability of artificial intelligence to fully compete against real players. The resulting system is functional, autonomous and capable of playing the game Scotland Yard on simplified game area. Based on these experiments, the thesis also introduces some improvements that could utilize machine learning and extend the existing solution.
|
guest ::
