Optimizacija neuronske mreže evolucijskim algoritmom za igru zmija

undergraduate thesis

• Title (english): Neural network optimisation for the snake game using evolutionary algorithms
• Author: Luka Terzić
• Mentor: Marko Đurasević (mentor)
• Committee member: Marko Đurasević (predsjednik povjerenstva)
• Committee member: Nikolina Frid (član povjerenstva)
• Committee member: Vjeko Kužina (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-07, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Ovaj rad se bavi razvojem agenta koji će bez inicijalnog znanja naučiti igrati igru zmija. Daje se opis rada i implementacije genetskog algoritma i umjetne neuronske mreže. Kao dio implementacije genetskog algoritma ostvareni su i svi genetski operatori. Prikazane su performanse rada umjetne neuronske mreže te se uspoređuju njeni rezultati s jednostavnim algoritmom najkraćeg puta. Razvijeno je grafičko korisničko sučelje kojim se može prikazati rad algoritma te usporediti performanse rada genetskog algoritma na različitim veličinama uređaja.

Abstract (english)

This thesis deals with the development of an agent which will learn how to play the game Snake without any initial knowledge of the game. The thesis also provides a description of genetic algorithms and artificial neural networks. All of the genetic operators have been developed as a part of the implementation. The performance of the artificial neural network is evaluated against a simple shortest path algorithm. A graphical user interface is developed which could be used for comparing the performance of the genetic algorithm on different screen sizes.

• Keywords: igra Zmija; genetski algoritam; genetski operatori; umjetna neuronska mreža; algoritam najkraćeg puta; grafičko korisničko sučelje
• Keywords (english): Snake game; genetic algorithm; genetic operators; artificial neural networks; shortest path algorithm; graphical user interface
• Language: croatian
• URN:NBN: urn:nbn:hr:168:108250
• 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: Closed access
• Created on: 2023-01-15 11:22:33