Abstract

Modern video games come with increasingly large and complex worlds to satisfy players’ demands for a rich and long-lasting playing experience. This development brings along new challenges: For instance, how can we design robust and believable characters that players can engage with in an open-ended way? Or how can we assess the quality of game content, especially when being procedurally generated? In this talk, Christian will motivate the use of intrinsically motivated reinforcement learning – a technique which currently gains strong momentum in the search for artificial general intelligence – to address the challenges of next-generation video games. He will give a comprehensive, interdisciplinary introduction to the concept of intrinsic motivation and motivate the design of computational models thereof. He will point out the opportunities and challenges such models hold for game AI, and highlight several applications from his research. The use of intrinsically motivated reinforcement learning for video game AI is still in its infancy, and Christian will consequently finish with a set of open questions and interesting research projects.

Short Bio:

Christian Guckelsberger is currently finishing his PhD in the Game AI Research Group at Queen Mary, University of London. Christian’s goal is to engineer autonomous artificial systems that would be deemed creative in their own right by unbiased observers, and he addresses this challenge with computational models of intrinsic motivation. Through both theoretical inquiry and applied studies in videogames, Christian demonstrates that such models can give rise to more general, robust and adaptive creative systems.