人間-エージェント間での読心ゲームにおける言い当て行動系列に依存したユーザーモデルの推定

概要

論文の詳細を見る
This paper proposes an adaptive interaction between a human and a life-like agent. We present the method to estimate the partner's context with the nth order MDP model. In this framework, an agent plays a mind reading game with a user in which the agent estimates the user's current mind state by the previous game context. Then the agent gradually learns to read the user's mind transition rules by estimating the transition probabilities of the nth order MDP model. To examine the regularity of the partner's context, we made the simple experiments to estimate a user's mind transition rules. As the experimental result, the number of correct guess of four subjects is much improved in the latter half of the game
一般社団法人情報処理学会の論文
2003-10-08