I am a Ph.D. candidate at School of Interactive Computing , Georgia Institute of Technology. I am working with Dr. Mark Riedl in the Entertainment Intelligence Lab, where we focus on making computers that can understand, author, administer, perform and present various forms of narratives in order to create entertaining experiences for a human audience.
Narrative intelligence, or the ability to create, tell and understand narratives, is an important cognitive capability. From ancient Greek plays to modern motion pictures, from bedtime stories to Nobel-winning novels, storytelling in various forms plays a pervasive role in our lives. Research suggests that narratives help us to communicate effectively, to establish identities, and to learn languages. Hence, narrative intelligence is an important component of machines aiming to simulate human intelligence or to communicate effectively with humans. However, existing computational attempts have been limited by the lack of narrative knowledge.
My dissertation focuses on learning complex knowledge structures that support computational narrative intelligence. In particular, I try to answer these research questions: Can a computational system acquire structured knowledge in support of narrative intelligence? What is a good computational representation for such knowledge? Can the learned knowledge be utilized efficiently in story creation, storytelling, and story understanding?
I strive to achieve three goals in my research: (1) understand the algorithms and mathematics well, but not restrain our thoughts to established formalism and assumptions, however theoretically well-formed they may be. (2) Not be afraid to learn about new fields because intersections often bring inspiration. (3) Find a balance between practicality and potential impact, as perfect can get in the way of good.