Interactive Super Mario Bros Evolution

Creating controllers for NPCs in video games is traditionally a challenging and time consuming task. Automated learning methods such as neuroevolution (i.e. evolving artificial neural networks) have shown promise in this context but they often require carefully designed fitness functions to encourage the evolution of desired behaviors. In this paper, we show how casual users can create controllers for \emph{Super Mario Bros} through an interactive evolutionary computation (IEC) approach, without prior domain or programming knowledge. By iteratively selecting Super Mario behaviors from a set of candidates, users are able to guide evolution towards a variety of different behaviors, which would be difficult with an automated approach. Additionally, the user-evolved controllers perform similarly well as controllers evolved with a traditional fitness-based approach when comparing distance traveled. The results suggest that IEC is a viable alternative in designing complex controllers for video games that could be extended to other games in the future.