| Project/Area Number |
17K12807
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Entertainment and game informatics 1
|
|---|
| Research Institution | Kochi University of Technology (2018-2021)
The University of Tokyo (2017) |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2017-04-01 – 2022-03-31
|
| Project Status |
Completed (Fiscal Year 2021)
|
| Budget Amount *help |
¥3,510,000 (Direct Cost: ¥2,700,000、Indirect Cost: ¥810,000)
Fiscal Year 2019: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2018: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2017: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
|
|---|
| Keywords | 助言 / 探索 / ゲーム / 協調 / ゲームプログラミング / ゲーム情報学 |
|---|
| Outline of Final Research Achievements |
Research and development of artificial intelligence is advancing, and AI with higher performance than humans can be obtained in some fields. On the other hand, there are areas where human intuition prevails, and a system in which humans and AI cooperate with each other will be important. In games such as shogi and the game of go, players can improve their playing ability by receiving advice. Therefore, we studied a system that utilizes "advice" as a cooperative system.
We showed that in shogi, game AI can improve its performance by receiving search results as advice from other game AI, comparing them with its own search results, and then search additionally. Furthermore, interesting results were obtained, such as the fact that the advisor does not necessarily need to be stronger than the player receiving the advice.
|
| Academic Significance and Societal Importance of the Research Achievements |
人間よりも性能の高いAI が多くの分野で使われるようになっている。人間とAI の協調が必要な場面もあり、そのシステムの一つとして助言システムが考えられる。AI の方が一般にミスは少ないことを考えると、人間からAI へと助言するようなシステムとなると考えれる。
そのようなシステムについて、将棋とそのゲームAI を題材として具体的なシステムの構築とその有効性について研究し、有効性を示すことができた。ゲームAI の強化という観点での成果が得られている他、助言者の性能が助言を受ける側より必ずしも優れている必要がないという結果は興味深く、ゲーム以外での助言システム利用に有用な知見であると考えられる。
|
