Minecraft; fitted-Q; reinforcement learning; extra-trees; video game; artificial intelligence; intelligent agent; large state space; FQI; jeu vidéo; intelligence artificielle; apprentissage
Abstract :
[en] Video games have significantly evolved since the emergence of first arcade games. They have
become more and more complex and now allow for the simulation of advanced environments.
This work focuses on the development of an intelligent agent that aims at collecting a given
type of ressource in random worlds from Minecraft, a well-known video game.
Disciplines :
Civil engineering
Author, co-author :
Lorent, Pierre ; Université de Liège - ULiège > 2e an. master ingé. civ. info., fin. appr.
Language :
English
Title :
Sequential decision making under uncertainty in randomly generated games : A Minecraft intelligent agent for resource gathering
Alternative titles :
[fr] Prise de décision séquentielle sous incertitude dans les jeux vidéos : Un agent intelligent pour récolter des ressources dans Minecraft
Defense date :
24 June 2014
Number of pages :
32
Institution :
ULiège - Université de Liège
Degree :
Master ingénieur civil en informatique à finalité approfondie
Promotor :
Ernst, Damien ; Université de Liège - ULiège > Montefiore Institute of Electrical Engineering and Computer Science
President :
Boigelot, Bernard ; Université de Liège - ULiège > Montefiore Institute of Electrical Engineering and Computer Science
Jury member :
Gemine, Quentin ; Université de Liège - ULiège > Montefiore Institute of Electrical Engineering and Computer Science
Fonteneau, Raphaël ; Université de Liège - ULiège > Montefiore Institute of Electrical Engineering and Computer Science
Geurts, Pierre ; Université de Liège - ULiège > Montefiore Institute of Electrical Engineering and Computer Science