Self Improving and Self Adaptive Systems - Exemplified by Intelligent Tutoring
In this 6-week lecture (total of 12 lectures, 2 hours each), I will briefly cover four aspects of the science of learning:
Theories: how people remember and learn.
Technologies: how to implement learning theories within working systems, such as Intelligent Tutoring Systems.
Applications: development of learning systems that enhance instruction in diverse areas, such as Mathematics, Physics, Reading, and Electronics.
Impacts: execution of efficacy studies of systems in real-world settings.
In the rest of the 6 weeks, I will introduce a general framework for the Adaptive Instructional System (AIS) (namely, Generalized Intelligent Framework for Tutoring (GIFT)) and a specific instance of AIS (namely AutoTutor, a conversation-based Intelligent Tutoring System (CbITS)), At the end of the course, I expect each of the students will be able to create a CbITS that is self-improvable with certain levels of self-improvability. We will examine the self-improving process for some of the examples produced by the students.