Računalni inteligentni igrač za igranje Zatvorenikove dileme

undergraduate thesis

• Title (english): Computer-based intelligent player for Prisoner's Dillema Game
• Author: Petar Belošević
• Mentor: Marko Čupić (mentor)
• Committee member: Marko Čupić (predsjednik povjerenstva)
• Committee member: Ivan Grubišić (član povjerenstva)
• Committee member: Stjepan Požgaj (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-07-04, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Zatvorenikova dilema je poznati problem iz područja teorije igara. U jezgri dileme
je sukob između racionalnog razmišljanja maksimiziranja vlastite dobiti dvaju igrača
koje ih dovodi do sub-optimalne pozicija te suradnje koja je korisna za oba igrača. Ovaj
rad se bavi problemom izrade inteligentnog igrača koji će moći igrati ponavljajući pro
blem Zatvorenikove dileme. Za modeliranje inteligentnog igrača korištena je Elmanova
neuronska mreža. Mreža je trenirana generacijskim genetskim algoritmom. Pokusom
je pokušano utvrditi učestalost pojedinih karakteristika strategije igranja inteligentnog
igrača. Timeseželiutvrditi robusnoststrategija igranja Zatvorenikove dilemekojeposje
duju tražene karakteristike. Karakteristike koje se promatraju su: ljubaznost, sklonost
opraštanju i provokabilnost.

Abstract (english)

Prisoner’s Dilemma is famous problem in game theory. The core of the Dilemma is
the conflict between rational thinking of maximizing personal gain that brings two play
ers in sub-optimal position and mutual cooperation that brings benefits for both players.
Thisthesis hasgoaltocreateintelligent player for iterated Prisoner’s Dilemma. Formod
elling an intelligent player was used Elman’s neural network. The network was trained
with generational genetic algorithm. Experiment was used to determine frequency of
specific characteristics in the strategies used by the intelligent player. Goal of the experi
ment is to show howrobust strategies that have specified characteristics are. Character
istics that were observed are: niceness, forgiveness and provocability.

• Keywords: Zatvorenikova dilema; umjetna inteligencija; umjetne neuronske mreže; inteligentni igrač; Elmanova neuronska mreža
• Keywords (english): Prisoner’s dilemma; Artificial intelligence; Artificial neural networks; Intelligent player; Elman’s neural network
• Language: croatian
• URN:NBN: urn:nbn:hr:168:054966
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2024-10-12 20:05:22