One important impact of technology on 21st century education has been the advances made possible by mining the wealth of user data from various learning technologies. In this talk, I will discuss my research on the application of machine Learning and data mining to improve human learning, and new challenges exposed by human learning. I will focus on the induction of effective pedagogical tutorial strategies via Reinforcement Learning (RL). These strategies guide the systems' fine-grained adaptations to the students' learning process. There is no well-established theory governing fine-grained adaptation and it has yet to be shown that either human or computer tutors have effective adaptive strategies. In this work, I demonstrated that RL can be applied to induce effective instructional policies from preexisting interaction data and that fine-grained strategies are a source of pedagogical power.

Dr. Min Chi is a Post-Doctoral Fellow in the Human-Sciences and Technologies Advanced Research Institute at Stanford University. She was previously a postdoctoral fellow in the Machine Learning Department in the School of Computer Science at Carnegie Mellon University. She received her Ph.D. from the Intelligent Systems Program at the University of Pittsburgh. Her research sits at the intersection of learning technology, machine learning & data mining, learning science and cognitive science. She has received the Best Paper Award at the Intelligent Tutoring Systems Conference and the James Chen Best Student Paper Award at the User Modeling, Adaptation and Personalization Conference in 2010. She also received the Best Student Paper Award at the 2008 Intelligent Tutoring Systems Conference