Designing Intelligent Interactive Systems from an Information-theoretic Perspective

In this talk, I explore the notion of information in the human-computer communication process and design intelligent interactive systems using the tools of information theory. Particularly, I propose BIG (Bayesian Information Gain), a framework to quantify the information sent by the user to the computer to express her intention. Two applications, BIGnav for multiscale navigation and BIGFile for hierarchical file retrieval, demonstrate how the computer can play a more active role and work together with the user to achieve shared goals. The third application, Entrain, also shows how the system can shape user experience in the context of collective music making (live demo).
My general research interest lies in using computational approaches to design intelligent interactive systems, particularly taking advantage of explicit and implicit information such as user’s intention, attention and semantic activities to empower interaction in artistic and nonartistic settings.