Apprentice Learner Architecture: A Framework for Modeling Human Learning from Demonstrations and Feedback in Digital Environments
Proceedings of the Students of Cognitive Systems Workshop at the Fourth Annual Conference on Advances in Cognitive Systems
2016
Abstract
Understanding the nature of human intelligence and developing intelligent agents capable of modeling humans are fundamental goals of cognitive systems research. Prior work modeling human problem solving has explored how hand-constructed domain models (e.g., production-rule models) can be used to explain human behavior. Typically, these models account for how humans improve their problem-solving performance given practice (i.e., speed-up learning), but they do not account for how humans acquire initial domain models. One approach that humans use to acquire knowledge in a new domain is apprenticeship learning, or learning from demonstrations and feedback from an expert. In the current work, I formalize the apprenticeship learning task for digital learning environments and present the Apprentice Learner Architecture, which provides a framework for building models of apprenticeship learning that align with this task formalization. Next, I briefly review how this model can be used to simulate and predicting human behavior in intelligent tutors. Finally, I conclude with directions for future work.
BibTeX
@inproceedings{maclellan-acs-2016-ws,
title = {Apprentice Learner Architecture: A Framework for Modeling Human Learning from Demonstrations and Feedback in Digital Environments},
author = {MacLellan, Christopher J.},
booktitle = {Proceedings of the Students of Cognitive Systems Workshop at the Fourth Annual Conference on Advances in Cognitive Systems},
year = {2016},
}