Institutional Repository of Vilnius University
Stalo žaidimo dirbtinio intelekto kūrimas naudojant NNUE ir Alfa-Beta paiešką /
Abstract
dc:descriptionThis paper describes hexagonal chess artificial intelligence development using Alpha-Beta pruning and an evaluation function based on efficiently updatable neural networks (NNUE). NNUE based evaluation function was compared to hand-crafted evaluation function, which was used to generate the training data. Two aspects of these evaluation functions were compared – speed and strength. Alpha-Beta pruning efficiency was also analysed at different search tree depths. NNUE based evaluation function speed is similar to hand-crafted evaluation function speed with 1 level shallower Alpha-Beta search tree. NNUE evaluation was twice as weak compared to hand-crafted evaluation function which was used to generate the training data. Alpha-Beta pruning is effective event at small search tree depths.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Jokūbauskas, Gabrielius,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD146241349&prefLang=en_US