2016 Fiscal Year Final Research Report
Enhancement method of game AI based on postmortem
| Project/Area Number |
26730181
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Entertainment and game informatics 1
|
|---|
| Research Institution | The University of Tokyo (2016)
Hokkaido University (2014-2015) |
|---|
Principal Investigator |
TAKEUCHI Shogo 東京大学, 大学院新領域創成科学研究科, 特任研究員 (40625258)
|
|---|
| Project Period (FY) |
2014-04-01 – 2017-03-31
|
| Keywords | 感想戦 / 協調 |
|---|
| Outline of Final Research Achievements |
In games such as shogi and game of go, human players discusses about their play for improving their skill. Such a discussion is called "Kanso-sen" (postmortem) and game AIs does not discuss about theire play. In this research, we focused on the "Kanso-sen" and try to improve AIs' performance by discussion between multiple game AIs.
One of the main achievements of this research is to propose a method for giving advice to another game AI to a certain game AI and to show improvement in cooperation by it.
|
| Free Research Field |
ゲーム情報学
|
