JAIST Repository >
b. 情報科学研究科・情報科学系 >
b11. 会議発表論文・発表資料等 >
b11-1. 会議発表論文・発表資料 >

このアイテムの引用には次の識別子を使用してください: http://hdl.handle.net/10119/13831

タイトル: Machine-Learning of Shape Names for the Game of Go
著者: Ikeda, Kokolo
Shishido, Takanari
Viennot, Simon
キーワード: Go
Machine Learning
Shape Name
Entertainment
発行日: 2015-12-25
出版者: Springer
誌名: Lecture Notes in Computer Science
巻: 9525
開始ページ: 247
終了ページ: 259
DOI: 10.1007/978-3-319-27992-3_22
抄録: Computer Go programs with only a 4-stone handicap have recently defeated professional humans. Now that the strength of Go programs is sufficiently close to that of humans, a new target in artificial intelligence is to develop programs able to provide commentary on Go games. A fundamental difficulty in this development is to learn the terminology of Go, which is often not well defined. An example is the problem of naming shapes such as Atari, Attachment or Hane. In this research, our goal is to allow a program to label relevant moves with an associated shape name. We use machine learning to deduce these names based on local patterns of stones. First, strong amateur players recorded for each game move the associated shape name, using a pre-selected list of 71 terms. Next, these records were used to train a supervised machine learning algorithm. The result is a program able to output the shape name from the local patterns of stones. Including other Go features such as change in liberties improved the performance. Humans agreed on a shape name with a rate of about 82 %. Our algorithm achieved a similar performance, picking the name most preferred by the humans with a rate of about 82 %. This performance is a first step towards a program that is able to communicate with human players in a game review or match.
Rights: This is the author-created version of Springer, Kokolo Ikeda, Takanari Shishido, and Simon Viennot, Lecture Notes in Computer Science, 9525, 2015, 247-259. The original publication is available at www.springerlink.com, http://dx.doi.org/10.1007/978-3-319-27992-3_22
URI: http://hdl.handle.net/10119/13831
資料タイプ: author
出現コレクション:b11-1. 会議発表論文・発表資料 (Conference Papers)

このアイテムのファイル:

ファイル 記述 サイズ形式
21682.pdf470KbAdobe PDF見る/開く

当システムに保管されているアイテムはすべて著作権により保護されています。

 


お問い合わせ先 : 北陸先端科学技術大学院大学 研究推進課図書館情報係