Interactive Explainable Case-Based Reasoning for Behavior Modelling in Videogames

Creation of believable characters is one of the most challenging problems in the video game industry. Although there are different authoring tools available for designers and programmers to create the behavior of non-player characters, it remains a complex and error prone process that requires a high level of technical knowledge. Considering how powerful Learning from Demonstration is for building intelligent agents that are able to replicate human player behaviors, we are currently using an online case-based reasoning agent that learns how to imitate real players of Ms. Pac-Man using an interactive approach in which both the human player and the computational agent take turns controlling the character. In this paper, we describe the new explanation subsystem and how it helps game designers to understand the agent’s learning process. We also present an evaluation of the system performed by five professional video game designers.