Title

LEVEL GENERATION USING GENETIC ALGORITHMS AND DIFFICULTY TESTING USING REINFORCEMENT LEARNING IN MATCH-3 GAME

Abstract

Abstract

Gaming industry is an enormous industry which contains game development, art, andmarketing games developed. It has grown faster in recent years, especially in mobilesection. Competition in mobile gaming increased and it brings us some concepts inthis area such as quick prototypes, automation, minimum viable product. In order tomaintain a successful job in mobile gaming industry, you need to minimize mistakesand long periods to create a game. Level generation is the one of the reasons why theperiod for development to product is long because unique and well-adjusted difficultyfor a level is generally tested by humans many times for one level to assure that thelevel is ready to be added. This thesis aims to come through those issues by automatedlevel generation for a match-3 game using genetic algorithms and testing all generatedlevels using reinforcement algorithms to minimize the time consumption for a leveldesigner. This will help them to easily and quickly generate many levels that theirdifficulties are already determined by using reinforcement learning.

Join Zoom Meeting
https://zoom.us/j/7593449920?pwd=SnRZLzYvK1lINDljbDFDNk9VS29tQT09
Meeting ID: 759 344 9920
Passcode: 753917

Supervisor(s)

Supervisor(s)

SAMET ALP DUKKANCI

Date and Location

Date and Location

2021-09-08 11:00:00

Category

Category

MSc_Thesis