• Apprentice Learning

    This research aims to develop a unified theory of learning from instruction and to provide a framework for building machines that are natural to teach. Central to this work is the Apprentice Learner Architecture, which faclitates model development and testing, and the Natural Training Interaction Framework, which outlines the space of natural training interactions.

  • Probabilistic Concept Formation

    This research line explores human-inspired models of concept formation. Unlike most ML work, which emphasizes batch processing over structure-less vectors, I emphasize the incremental acquisition of structured concepts from a continuous stream of experiences.

  • Learner Modeling

    This line of work centers around descriptive modeling of learners and their behavior within educational technologies, such as tutoring systems and educational games. Key projects in this line include accounting for “slipping” (accidental errors) and strategy modeling.

  • Conceptual Design Aids

    This line of work aims to better understand problem formuation and early-stage conceptual design processes in humans. The ultimate goal of this work is to construct tools that facilitate creative design. Recent work in this line has explored how to assess design creativity at scale using crowdsourcing and statistical modeling.