2016 Fiscal Year Final Research Report
Large scale distributed monte-carlo game-tree search based on probability distribution
| Project/Area Number |
26280130
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Partial Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Entertainment and game informatics 1
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Renkei-kenkyūsha) |
TSURUOKA Yoshimasa 東京大学, 工学(系)研究科(研究院), 准教授 (50566362)
|
|---|
| Project Period (FY) |
2014-04-01 – 2017-03-31
|
| Keywords | 人工知能 / アルゴリズム / ゲーム情報学 |
|---|
| Outline of Final Research Achievements |
Large-scale search problems in real world are not applicable exhaustive search; randomized search algorithms have great ability to explore such problems. Game tree search is an example of such problem; Monte-Carlo Tree Search algorithm (MCTS) has been developed and is getting widely used with good performance. This great advance, however, does not help to achieve good performance in Shogi that has long narrow path of 'correct' play.
We propose a new randomized game-tree search algorithm based on Bayesian Approach that can treat evaluated values as probability distributions. In this research we evaluate the effectiveness of our approach using distributed computing method.
|
| Free Research Field |
ゲーム情報学
|
