The ability to estimate human behavior state is not sufficient for building cooperative agents. Predicting how people adapt their behavior in response to an agent's actions is also necessary. We propose a new approach based on computational rationality, which combines reinforcement learning and cognitive modeling to facilitate machine understanding of humans.
The ability to estimate the state of a human partner is an insufficient basis on which to build cooperative agents. Also needed is an ability to predict how people adapt their behavior in response to an agent's actions. We propose a new approach based on computational rationality, which models humans based on the idea that predictions can be derived by calculating policies that are approximately optimal given human-like bounds. Computational rationality brings together reinforcement learning and cognitive modeling in pursuit of this goal, facilitating machine understanding of humans.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据