prikaz prve stranice dokumenta Predviđanje ishoda šahovske igre korištenjem rekurentnih neuronskih mreža
  Download
PDF 17.15 MB
master's thesis
Predviđanje ishoda šahovske igre korištenjem rekurentnih neuronskih mreža
Osijek: Josip Juraj Strossmayer University of Osijek, Department of Mathematics, 2023. urn:nbn:hr:126:630671
Vuković, Filip
Josip Juraj Strossmayer University of Osijek
Department of Mathematics

Institutional repository: Repository of School of Applied Mathematics and Computer Science

Cite this document

Vuković, F. (2023). Predviđanje ishoda šahovske igre korištenjem rekurentnih neuronskih mreža (Master's thesis). Osijek: Josip Juraj Strossmayer University of Osijek, Department of Mathematics. Retrieved from https://urn.nsk.hr/urn:nbn:hr:126:630671

Vuković, Filip. "Predviđanje ishoda šahovske igre korištenjem rekurentnih neuronskih mreža." Master's thesis, Josip Juraj Strossmayer University of Osijek, Department of Mathematics, 2023. https://urn.nsk.hr/urn:nbn:hr:126:630671

Vuković, Filip. "Predviđanje ishoda šahovske igre korištenjem rekurentnih neuronskih mreža." Master's thesis, Josip Juraj Strossmayer University of Osijek, Department of Mathematics, 2023. https://urn.nsk.hr/urn:nbn:hr:126:630671

Vuković, F. (2023). 'Predviđanje ishoda šahovske igre korištenjem rekurentnih neuronskih mreža', Master's thesis, Josip Juraj Strossmayer University of Osijek, Department of Mathematics, accessed 16 June 2024, https://urn.nsk.hr/urn:nbn:hr:126:630671

Vuković F. Predviđanje ishoda šahovske igre korištenjem rekurentnih neuronskih mreža [Master's thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Department of Mathematics; 2023 [cited 2024 June 16] Available at: https://urn.nsk.hr/urn:nbn:hr:126:630671

F. Vuković, "Predviđanje ishoda šahovske igre korištenjem rekurentnih neuronskih mreža", Master's thesis, Josip Juraj Strossmayer University of Osijek, Department of Mathematics, Osijek, 2023. Available at: https://urn.nsk.hr/urn:nbn:hr:126:630671

Metadata
TitlePredviđanje ishoda šahovske igre korištenjem rekurentnih neuronskih mreža
Title (english)Chess Game Outcome Prediction Based on Reccurent Neural Networks
AuthorFilip Vuković
MentorDomagoj Matijević (mentor)
Committee memberDomagoj Ševerdija (predsjednik povjerenstva)
Committee memberDomagoj Matijević (član povjerenstva)
Committee memberKristian Sabo (član povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
Department of Mathematics
Osijek
Defense date and country2023-05-18, Croatia
Scientific / art field,
discipline and subdiscipline		NATURAL SCIENCES
Mathematics
Mathematical Logic and Accounting
Abstract
U ovom radu bavimo se kreiranjem neuronskog modela za predviđanje ishoda šahovske igre. Za razumijevanje neuronskih modela, prvo definiramo jednostavne neuronske mreže te uvodimo različite varijacije i metode korištene u području strojnog učenja. Potom, definiramo rekurentne neuronske mreže i procese potrebne za učenje takvih mreža. Osnovnu strukturu rekurentnih mreža nadograđujemo s LSTM arhitekturom potrebnu za definiranje modela projektnog zadatka. Nadalje, prikazujemo način zapisivanja šahovskih igara u standardnoj notaciji i uvodimo dva načina prikazivanja stanja šahovske ploče pogodnih za korištenje u neuronskim modelima. Na kraju, definiramo model temeljen na LSTM arhitekturi koji obrađuje šahovsku igru putem niza poteza i uspoređujemo rezultate modela u odnosu na dvije reprezentacije šahovske ploče.
Abstract (english)
In this paper, we focus on creating a neural model for predicting the outcome of a game of chess. To understand neural models, we first define simple neural networks and introduce various variations and methods used in the field of machine learning. Then, we define recurrent neural networks and the processes required for learning such networks. We enhance the basic structure of recurrent networks with the LSTM architecture necessary for defining the project task model. Furthermore, we show how to record chess games in standard notation and introduce two ways of representing the state of the chessboard suitable for use in neural models. Finally, we define a model based on the LSTM architecture that processes a chess game through a sequence of moves and compare the results of the model with respect to two representations of the chessboard.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:126:630671
Study programmeTitle: Mathematics; specializations in: Financial and Statistical Mathematics, Mathematics and Computer Science, Industrial and Applied Mathematics
Course: Financial and Statistical Mathematics
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra matematike (magistar/magistra matematike)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2023-05-19 11:37:21