Development of shogi-endgame algorism exceeding the Meijin
| Project/Area Number |
20700133
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | Matsue National College of Technology (2009-2011)
Japan Advanced Institute of Science and Technology (2008) |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2008 – 2011
|
| Project Status |
Completed (Fiscal Year 2011)
|
| Budget Amount *help |
¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2011: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2010: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2009: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2008: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
|
|---|
| Keywords | コンピュータ将棋 / AND/OR木探索 / 終盤アルゴリズム / 詰将棋探索 / 将棋終盤 / Weak proof number search / 寄せ合い / Bonanza Method / AND / OR木探索 / UCT / モンテカルロ木探索 / 機械学習 / 囲碁 / オセロ / 詰み探索 / proof branching number / 熟達化 / 手筋 / n-gram / 詰め将棋 / Weak Proof Number Search / 証明数探索 / 評価関数学習 / コンピュータ囲碁 |
|---|
| Research Abstract |
"Weak Proof Number Search"(WPNS) which solves the double count problem, a difficult problem of AND/OR tree search, was developed. In experiments using tsume-shogi problems, the performance of WPNS overwhelmed others especially for super-difficult problems which exceeds hundreds plies. Moreover, evaluation function machine learning of the game of go or shogi and cognitive science research of shogi skill formation were done.
|
Report
(6 results)
Research Products
(38 results)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
[Presentation] を用いた囲碁評価関数の設計2009
Author(s)
橋本剛, 松井利樹, 野口陽来, Bonanza Method
Organizer
組合せゲーム・パズル ミニプロジェクト第4回ミニ研究集会
Place of Presentation
東京工業大学
Year and Date
2009-03-03
Related Report
-
-
-
-
-
-
-
-
-
-
