This paper studies a novel model-based method for artificial teaching; that is, the problem of selecting teaching interventions in interaction with humans. Previous work has either focused on the individualization of teaching or the optimization of teaching intervention sequences. We converge these two lines of research in an individualized model-based planning approach. In model-based planning, a user memory model's parameters are learned interactively and used to pick best interventions. New to our approach is the use of a model that can account for some key individual and material-specific characteristics related to recall/forgetting, along with a planning technique that considers users' practice schedules. Using a rule-based approach as a baseline, we evaluate the benefits of this approach in a controlled study of artificial teaching in second language vocabulary learning.
|Publication status||Accepted/In press - 2021|
|MoE publication type||Not Eligible|