My primary research is in artificial intelligence, a field that seeks to re-create the intelligent abilities that humans have developed. That goal has proved to be more challenging than the early researchers imagined, because there's a great deal we don't yet understand about human intelligence. Still, AI research has produced many fascinating and useful ideas.
I am also interested in the scientific study of education. In this kind of research, students and courses themselves are the subjects, and the goal is to develop more effective techniques for teaching and learning.
My graduate work focused on relational transfer in reinforcement learning. This former website describes what that means and lists my publications in that area.
At St. Lawrence, my projects have spanned a wider range of topics. The papers below are peer-reviewed publications that have sprung from recent work.
- The Turing Test in the Classroom
- Reinforcement learning agents providing advice in complex video games
Connection Science (2014)
- Teaching on a budget: agents advising agents in reinforcement learning
Conference on Autonomous Agents and Multiagent Systems (2013)
- Teaching problem-solving in algorithms and AI
Symposium on Educational Advances in Artificial Intelligence (2012)
- Towards student/teacher learning in sequential decision tasks
Conference on Autonomous Agents and Multiagent Systems (2012)
- Help an agent out: student/teacher learning in sequential decision tasks
Adaptive Learning Agents Workshop at AAMAS (2012)
- Lightweight adaptation in model-based reinforcement learning
Lifelong Learning Workshop at AAAI (2011)
- Student interest and choice in programming assignments
Northeastern Consortium for Computing Sciences in Colleges (2011)
- Crowd simulation via multi-agent reinforcement learning
Conference on Artificial Intelligence and Interactive Digital Entertainment (2010)