Razvoj strategije za igranje šaha

undergraduate thesis

• Title (english): Development of a strategy for playing chess
• Author: Josip Kelava
• Mentor: Marko Đurasević (mentor)
• Committee member: Marko Đurasević (predsjednik povjerenstva)
• Committee member: Lucija Planinić (član povjerenstva)
• Committee member: Nikolina Frid (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-07-06, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Igre su odličan problem za razvoj, učenje i inovaciju umjetne inteligencije. Šah je jedna od prvih poznatijih igri za koju je rađena naprednija umjetna inteligencija, te je napravljena i u sklopu ovog rada. Šah je igra na ploči između dva igrača koji naizmjenično igraju poteze. Umjetna inteligencija za igranje šaha implementirana je kao algoritam minmax, koji simulira sve moguće scenarije odigravanja određenog broja poteza, ocjenjuje dobivene pozicije te odabire najbolji potez pretpostavljajući da će protivnik igrati savršeno. Procjena pozicije implementirana je kao razlika ukupne vrijednosti figura na ploči oba igrača, te njihovih pozicija na ploči. Na tu procjenu dodane su evaluacije neuronske mreže trenirane genetskim algoritmom, poboljšavajući procjenu. Kako bi se osigurale raznolikost i općenitost zahtjeva genetskog algoritma, populacija je podijeljena u više manjih grupa s nemogućnosti prijenosa DNA iz jedne u drugu grupu.

Abstract (english)

Games are a great problem for development, study, and innovation in the field of artificial intelligence. Chess is one of the first more well-known games that had an advanced AI be made for, and one was made for this paper. Chess is a board game for two players who alternate making moves. Artificial intelligence for chess was implemented as a minmax algorithm, which simulates all possible scenarios of playing a fixed number of moves, assesses the reached positions and chooses the best move assuming that its opponent is playing perfectly. The assessment is implemented as the difference of values of all the pieces on the board of both players, and their positions on the board. Added to these assessments are neural network’s evaluations of the same positions, improving the assessment quality. This network can be trained using the genetic algorithm. To improve variance and generality of demands of the genetic algorithm, its population is spread into several smaller groups, preventing the spread of a single DNA from one group to another.

• Keywords: šah; umjetna inteligencija; minmax; genetski algoritam; neuoronska mreža
• Keywords (english): chess; artificial intelligence; minmax; genetic algorithm; neural network
• Language: croatian
• URN:NBN: urn:nbn:hr:168:670305
• 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: 2023-01-14 09:05:44