| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

Show document Help

Title:Razvoj in učenje agentov z okrepitvijo za računalniško igro Suika Game : diplomsko delo
Authors:ID Glavič, Nik (Author)
ID Potočnik, Božidar (Mentor) More about this mentor... New window
Files:.pdf VS_Glavic_Nik_2025.pdf (2,23 MB)
MD5: 1C1C680BF89D413C73182F2225AD8BB4
 
Language:Slovenian
Work type:Bachelor thesis/paper
Typology:2.11 - Undergraduate Thesis
Organization:FERI - Faculty of Electrical Engineering and Computer Science
Abstract:V diplomskem delu smo razvili okolje računalniške igre Suika Game in agenta, ki se igre uči igrati s pomočjo okrepitvenega učenja. Igro smo poustvarili v programskem okolju Unity. Za učenje agenta smo uporabili knjižnico ML-Agents. Raziskali smo temeljna načela okrepitvenega učenja ter preučili učinkovitost agenta pri igranju igre. Naš cilj je bil razviti strategijo, s katero agent postopoma izboljšuje svojo igro in dosega čim boljši rezultat. Na osnovi več eksperimentov smo ugotovili, da naučen agent pri igranju računalniške igre Suika Game dosega rezultate, primerljive ali celo boljše od človeškega igralca.
Keywords:nevronska mreža, videoigra, okrepitveno učenje, okolje Unity, knjižnica ML-Agents
Place of publishing:Maribor
Place of performance:Maribor
Publisher:[N. Glavič]
Year of publishing:2025
Number of pages:1 spletni vir (1 datoteka PDF (X, 40 str.))
PID:20.500.12556/DKUM-94996 New window
UDC:004.8:004.96(043.2)
COBISS.SI-ID:257486083 New window
Publication date in DKUM:15.10.2025
Views:115
Downloads:9
Metadata:XML DC-XML DC-RDF
Categories:KTFMB - FERI
:
Copy citation
  
Average score:(0 votes)
Your score:Voting is allowed only for logged in users.
Share:Bookmark and Share



Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Licences

License:CC BY-NC-ND 4.0, Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Link:http://creativecommons.org/licenses/by-nc-nd/4.0/
Description:The most restrictive Creative Commons license. This only allows people to download and share the work for no commercial gain and for no other purposes.
Licensing start date:01.09.2025

Secondary language

Language:English
Title:Development and training of reinforcement learning agents for the Suika computer Game
Abstract:In our thesis, we developed a Suika Game like environment and an agent that learns to play the game using reinforcement learning. The game was recreated in the Unity development environment, while the agent was trained with the ML-Agents library. We explored the fundamental principles of reinforcement learning and examined the agent’s performance in playing the game. Our objective was to design a strategy that enables the agent to gradually improve its gameplay and achieve the highest possible score. Based on multiple experiments, we found that the trained agent achieves results in Suika Game that are comparable to, or even better than, those of a human player.
Keywords:neural network, puzzle computer game, reinforced learning, Unity environment, ML-Agents library


Comments

Leave comment

You must log in to leave a comment.

Comments (0)
0 - 0 / 0
 
There are no comments!

Back
Logos of partners University of Maribor University of Ljubljana University of Primorska University of Nova Gorica