Primjena umjetne inteligencije u računalnoj igri Pacman

master's thesis

• Title (english): Application of AI in Computer Game Pacman
• Author: Krešimir Dodig
• Mentor: Tomislav Burić (mentor)
• Committee member: Tomislav Burić (predsjednik povjerenstva)
• Committee member: Ilko Brnetić (član povjerenstva)
• Committee member: Mate Puljiz (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-14, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Testiranje umjetne inteligencije u računalnim igrama je dobar pokazatelj koliko je ona zapravo napredovala. U ovom radu bavimo se igrom Pacman jer se ona pokazala kao iznimno zahtjevna za algoritme umjetne inteligencije. Za učenje i svladavanje igre Pacman, koristili smo varijaciju Q-učenja uz pristup koji zovemo Hibridna Arhitektura Nagrada (HRA). Ideja HRA je da rastavimo problem na manje probleme te imamo posebnog agenta za učenje za svaki od tih manjih problema. Ovaj pristup se pokazao odličan za Pacmana te smo uspješno svladali rješavanje (pobjeđivanje) igre.

Abstract (english)

Testing artificial intelligence in computer games is a good indicator of how far it has actually progressed. In this paper, we deal with the Pacman game because it has proven to be extremely demanding on artificial intelligence algorithms. To learn and master the Pacman game, we used a variation of Q-learning with an approach we call Hybrid Reward Architecture (HRA). The idea of HRA is to break down a problem into smaller problems and we have a special learning agent for each of those smaller problems. This approach proved to be great for Pacman and we successfully mastered solving (winning) the game.

• Keywords: Umjetna inteligencija; Pacman; Podržano učenje; Q-učenje; Hibridna Arhitektura Nagrada (HRA); Python
• Keywords (english): Artificial intelligence; Pacman; Reinforcement learning; Q-learning; Hybrid Reward Architecture (HRA); Python
• Language: croatian
• URN:NBN: urn:nbn:hr:168:416134
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 09:55:39