Automated Game Generation from Gameplay Videos

Faculty: 
Mark O. Riedl
Students: 
Matthew Guzdial

Game design is a difficult and costly practice. We developed a machine learning-based method for automatically generating games. We feed our system a series of gameplay videos of different games. Our system then leverages a technique we've developed called conceptual expansion to recombine its knowledge of game design to produce novel games.

You can see example videos of two output games here and here.

Lab: 
Faculty: 
Mark Riedl

The Entertainment Intelligence Lab focuses on computational approaches to creating engaging and entertaining experiences. Some of the problem domains they work on include, computer games, storytelling, interactive digital worlds, adaptive media and procedural content generation. They expressly focus on computationally "hard" problems that require automation, just-in-time generation, and scalability of personalized experiences.