Duboko učenje za procjenu kvalitete šahovskih pozicija

master's thesis

• Title (english): Deep Learning for Evaluation of Chess Positions
• Author: Luka Giljanović
• Mentor: Vladimir Čeperić (mentor)
• Committee member: Vladimir Čeperić (predsjednik povjerenstva)
• Committee member: Mirko Poljak (član povjerenstva)
• Committee member: Tomislav Marković (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-15, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Cilj ovog rada bio je istražiti kako evaluirati šahovsku poziciju pomoću dubokog učenja. Na
početku rada iznesena je problematika evaluacije šahovske pozicije. U prvom dijelu rada
opisuje se kako su se prikupljali podatci koji su kasnije korišteni u izgradnji modela. U
drugom dijelu rada opisano je kako su se podatci pripremili, odnosno kako su se prilagodili
za učenje modela konvolucijske neuronske mreže. U trećem dijelu rada prikazana je
izgradnja modela i objašnjena je njegova arhitektura. Konačno, prikazani su rezultati
testiranja modela.

Abstract (english)

The aim of this paper was to explore how to evaluate a chess position using deep learning.
At the beginning of the paper, the issue of evaluation of the chess position was presented.
The first part of the paper describes how the data, that were later used in building the model,
were collected. The second part of the paper describes how the data were prepared, ie how
they were adapted for convolutional neural network model. The third part of the paper
presents the construction of the model and explains its architecture. Finally, the test results
of the model are presented.

• Keywords: šah; evaluacija pozicije; duboko učenje; konvolucijske neuronske mreže; unaprijedne neuronske mreže; skup podataka; PGN notacija; FEN notacija
• Keywords (english): chess; position evaluation; deep learning; convolutional neural network; feedforward neural network; dataset; PGN notation; FEN notation
• Language: croatian
• URN:NBN: urn:nbn:hr:168:938811
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-14 08:48:41