Open AccessDeep Learning and the Game of Checkers
Author(s) -
Jan Popic,
Borko Bošković,
Janez Brest
Publication year - 2021
Publication title -
mendel ... (brno. on-line)/mendel ...
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.221
H-Index - 13
eISSN - 1803-3822
pISSN - 1803-3814
DOI - 10.13164/mendel.2021.2.001
Subject(s) - pruning , computer science , artificial neural network , game tree , monte carlo tree search , artificial intelligence , tree (set theory) , set (abstract data type) , machine learning , deep neural networks , theoretical computer science , sequential game , game theory , monte carlo method , mathematics , programming language , mathematical analysis , statistics , mathematical economics , agronomy , biology
In this paper we present an approach which given only a set of rules is able to learn to play the game of Checkers. We utilize neural networks and reinforced learning combined with Monte Carlo Tree Search and alpha-beta pruning. Any human influence or knowledge is removed by generating needed data, for training neural network, using self-play. After a certain number of finished games, we initialize the training and transfer better neural network version to next iteration. We compare different obtained versions of neural networks and their progress in playing the game of Checkers. Every new version of neural network represented a better player.