Intaractive Optimization Method based on Bidirectional Reinforcement Learning of Agent and Trainer and Verification with Practical Data
Project/Area Number |
17K00345
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Research Field |
Soft computing
|
Research Institution | Kyushu Institute of Technology |
Principal Investigator |
HORIO KEIICHI 九州工業大学, 大学院生命体工学研究科, 教授 (70363413)
|
Co-Investigator(Kenkyū-buntansha) |
磯貝 浩久 九州産業大学, 人間科学部, 教授 (70223055)
|
Project Period (FY) |
2017-04-01 – 2020-03-31
|
Project Status |
Completed (Fiscal Year 2019)
|
Budget Amount *help |
¥4,680,000 (Direct Cost: ¥3,600,000、Indirect Cost: ¥1,080,000)
Fiscal Year 2019: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2018: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2017: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
|
Keywords | メンタル状態 / フィードバックコメント / 強化学習 / 報酬設計 / 双方向強化学習 / エージェント / トレーナ / 認知誤差 / 学習過程の特徴量 / 分類 / 適切な報酬 / 特性分類 / 報酬最適化 / 行動変容 / 介入 |
Outline of Final Research Achievements |
In this research, we assume that the target agent acquires the behavior based on reinforcement learning, classify the characteristics of the agent, and design an appropriate reward method accordingly. First, in order to classify the characteristics of agents, we developed an application that measures the mental status of athletes and analyzed the relationship between the mental status of athletes and performance based on the collected data. In addition, as an appropriate way to give rewards to each player, we defined positive patterns, candid patterns, etc. regarding comments that are fed back to the players via the application, and confirmed that the tendency of appropriate feedback comments differs depending on the players.
|
Academic Significance and Societal Importance of the Research Achievements |
ヒトが社会で生活を営むにおいて,他者からの介入は非常に重要な意味を持つ.適切な介入はヒトのメンタル状態を改善し,結果としてその行動を変えることができる.本研究は,多数のスポーツ選手を例として,選手のメンタル状態とパフォーマンスの関係性を解析することで,各選手が良いパフォーマンスを発揮するメンタル状態を推定し,また,選手への介入として状態を解析結果を様々なパターンで提示することで,各選手にとって適切なコメントパターンを推定することを実現した.これらは,スポーツ現場のみならず,教育を始めとする様々なシーンに適用可能である.
|
Report
(4 results)
Research Products
(9 results)