Javascript must be enabled for the correct page display

Using a Support Vector Machine to learn to play Othello

Karavolos, K.D. (2010) Using a Support Vector Machine to learn to play Othello. Bachelor's Thesis, Artificial Intelligence.

[img]
Preview
Text
Danil_Karavolos_KIB_2010.pdf - Published Version

Download (338kB) | Preview

Abstract

Like chess and backgammon, the game Othello is a popular eld of application for Machine Learning (ML) techniques. An especially e ffective technique for learning to play a game is a neural network with Temporal Dif ference Learning (TDL). Such a neural netwerk has achieved a worldclass level of play in backgammon. Another succesful, increasingly popular ML technique is the Support Vector Machine (SVM). However, this technique has not yet been applied to a game. This experiment compares the use of an SVM to learn to play Othello with the use of a TD neural network and a few other techniques for playing Othello. It appears that the player that is trained with an SVM performs better than a player with random moves. However, it is generally defeated by the heuristic positional strategy and loses almost every game versus the mobility strategy and a TD player.

Item Type: Thesis (Bachelor's Thesis)
Degree programme: Artificial Intelligence
Thesis type: Bachelor's Thesis
Language: English
Date Deposited: 15 Feb 2018 07:30
Last Modified: 15 Feb 2018 07:30
URI: http://fse.studenttheses.ub.rug.nl/id/eprint/9068

Actions (login required)

View Item View Item