|
JAIST Repository >
School of Information Science >
Articles >
Journal Articles >
Please use this identifier to cite or link to this item:
https://hdl.handle.net/10119/12799
|
| Title: | Efficiency of Static Knowledge Bias in Monte-Carlo Tree Search |
| Authors: | Ikeda, Kokolo
Viennot, Simon |
| Keywords: | Monte Carlo Tree Search
Search Bias
Static Evaluation Function
Progressive Widening
Game of Go |
| Issue Date: | 2014-07-12 |
| Publisher: | Springer |
| Magazine name: | Lecture Notes in Computer Science |
| Volume: | 8427 |
| Start page: | 26 |
| End page: | 38 |
| DOI: | 10.1007/978-3-319-09165-5_3 |
| Abstract: | Monte-Carlo methods are currently the best known algorithms for the game of Go. It is already known that Monte-Carlo simulations based on a probability model containing static knowledge of the game are more efficient than random simulations. Such probability models are also used by some programs in the tree search policy to limit the search to a subset of the legal moves or to bias the search, but this aspect is not so well documented. In this article, we try to describe more precisely how static knowledge can be used to improve the tree search policy, and we show experimentally the efficiency of the proposed method with a large number of games against open source Go programs. |
| Rights: | This is the author-created version of Springer, Kokolo Ikeda and Simon Viennot, Lecture Notes in Computer Science, 8427, 2014, 26-38. The original publication is available at www.springerlink.com, http://dx.doi.org/10.1007/978-3-319-09165-5_3 |
| URI: | https://hdl.handle.net/10119/12799 |
| Material Type: | author |
| Appears in Collections: | b10-1. 雑誌掲載論文 (Journal Articles)
|
Files in This Item:
| File |
Description |
Size | Format |
|---|
| 19519.pdf | | 108Kb | Adobe PDF | View/Open |
|
All items in DSpace are protected by copyright, with all rights reserved.
|
